Aalto University

School of Science

Degree Programme in Industrial Engineering and Management

Alexander Gerbov

Process improvement and BIM in infrastructure

design projects – findings from 4 case studies in

Finland

Master’s Thesis

Espoo, 19/11/2014

Supervisor: Professor Vishal Singh

Instructors: Ph.D. Maila Suvanto, M.Sc. Heikki Halttula

II

AALTO UNIVERSITY SCHOOL OF SCIENCE ABSTRACT OF THE

Degree Programme in Industrial Engineering and Management MASTER´S THESIS

Author: Alexander Gerbov

Subject of the thesis: Process improvement and BIM in infrastructure design projects – findings from 4 case studies in Finland

Number of pages: 105

Date: 19.11.2014

Library location: TU

Professorship: Computer Integrated Construction

Code of professorship: -

Supervisor: Professor Vishal Singh

Instructors: Ph.D. Maila Suvanto, M.Sc. Heikki Halttula

Building information modelling is increasingly adopted in today’s

infrastructure sector. However, the implementation of BIM in the design

stage of construction projects and the benefits of BIM for design stage are

not sufficiently studied. The adoption of BIM may call for changes in

traditional design process. This research explores the current state of BIM

implementation and design practice, and the possibilities for process

improvement in the design phase of transport infrastructure lifecycle, by

studying four case design projects in Finland.

The data has been obtained in eight workshops with the members of the case

project teams; in addition, preliminary interviews and a questionnaire survey

have been conducted. The major revealed benefits of BIM are clash

detection, cost estimation and machine control. However, the potential

process benefits of BIM for design are not achieved. The study reveals a gap

between BIM process and design process in the projects, and a knowledge

gap between BIM experts and common designers. The major identified

barriers towards achieving the higher level of BIM implementation are the

inefficient information exchange, the lack of BIM skills and learning policies,

and the lack of modelling guidelines and procedures.

In addition, the work discusses the difficulties that have been encountered

when attempting to carry out quantitative measurements in design projects,

and proposes recommendations for those who plan to conduct such studies

in future.

Keywords: BIM, infrastructure, design, information modelling, collaboration

Publishing language: English

III

Acknowledgements

This Master’s thesis research would not have been possible without the involvement

and support of many people. At the end of this interesting and at times difficult journey,

I would like to express gratitude to everyone who helped me.

First of all, I would like to thank my supervisor Professor Vishal Singh, whose

experience and advice helped me enormously throughout the project. I would like to

thank the case companies and the participants of the workshop sessions, whose names

I cannot mention here, for their collaboration and commitment during the project. The

possibility to conduct the empirical part of my thesis was provided to me by Vianova

Systems Finland Oy, along with the financial support and a wonderful place to work.

I am very grateful for the continuous involvement and help of Maila Suvanto from

Vianova; without her help it would have been so much harder to proceed in this

research. I would like to thank Heikki Halttula, the managing director at Vianova

Systems Finland Oy, for making this research possible and for direct involvement and

advice. I also want to thank all the employees of the company for their friendliness,

which made my work there an enjoyable experience.

I am grateful to Professor Jan Holmström from my department who helped me to find

such a good opportunity for my thesis project, and helped me throughout my studies.

Last but not least, I express deepest gratitude to my parents for their love and for

supporting me in my studies.

Alexander Gerbov,

15.11.2014

IV

Table of contents

1. Introduction .............................................................................................................. 1

1.1. Background of the study .................................................................................... 1

1.2. Research scope and objectives .......................................................................... 3

1.3. Research motivation .......................................................................................... 4

1.4. Research planning and the literature study ........................................................ 7

2. Theoretical foundations ............................................................................................ 9

2.1. Overview of Building Information Modelling .................................................. 9

2.2. Ontology of collaborative design .................................................................... 14

2.3. Application of Lean principles to collaborative design ................................... 18

2.4. The characteristics of collaborative design in project management ................ 21

2.5. Benefits of technology for collaborative design in construction sector .......... 27

2.6. Conclusions and introduction to empirical study ............................................ 30

3. Methodology .......................................................................................................... 31

3.1. Case studies ..................................................................................................... 31

3.2. Sentiment analysis ........................................................................................... 34

3.3. Questionnaire ................................................................................................... 37

4. Results of the empirical study ................................................................................ 38

4.1. Questionnaire study ......................................................................................... 38

4.2. Case project A ................................................................................................. 44

4.2.1. Description ............................................................................................... 44

4.2.2. Workshop discussion ................................................................................ 46

4.2.3. Problems of design projects, Team 1 ....................................................... 51

4.3. Case project B ................................................................................................. 53

4.3.1. Description ............................................................................................... 53

4.3.2. Workshop discussion ................................................................................ 53

V

4.3.3. General problems and the ideal project – Team 2 .................................... 59

4.4. Case projects X and Y ..................................................................................... 62

4.4.1. Description ............................................................................................... 62

4.4.2. Workshop discussion ................................................................................ 63

4.4.3. General problems in design projects – Team 3 ........................................ 69

5. Discussion and Implications ................................................................................... 71

5.1. The current state of infrastructure design in the case projects ........................ 72

5.1.1. Design process in the case projects .......................................................... 72

5.1.2. Building Information Modelling in the case projects ............................... 79

5.2. Towards BIM-based collaborative design in construction .............................. 86

5.2.1. BIM as a tool for collaborative design ..................................................... 87

5.2.2. Barriers for achieving BIM-based collaboration in design projects ......... 90

5.3. Summary and practical recommendations ...................................................... 91

5.4. Validation of the results and recommendations for future research ................ 95

5.5. Exploring the possibility for quantitative research on AEC design activities . 97

References ................................................................................................................ 101

Appendix: The questionnaire form .......................................................................... - 1 -

VI

List of figures

Figure 1: The research framework ............................................................................... 5

Figure 2: Influence of overall project cost over the project lifecycle .......................... 5

Figure 3: The four components of organisation’s performance: people, process,

technology and policy .................................................................................................. 6

Figure 4: The research process ..................................................................................... 9

Figure 5: The conceptual framework of collaborative design system in professional

domains ...................................................................................................................... 18

Figure 6: Implementation steps to develop a computer program for delivery to a

customer ..................................................................................................................... 21

Figure 7: Cost of changes for design services ............................................................ 22

Figure 8: Parallel set narrowing process sketched by a Toyota manager .................. 25

Figure 9: Examples of note templates ........................................................................ 33

Figure 10: Fishbone (Ishikawa) diagram .................................................................... 34

Figure 11: Data collected in the empirical study ........................................................ 38

Figure 12: The willingness to work in projects that use BIM, in relation to BIM

experience. .................................................................................................................. 39

Figure 13: The level of satisfaction with the process of working with BIM in relation

to BIM experience. ..................................................................................................... 39

Figure 14: The perception of the effect of BIM on different aspects of designing, in

relation to BIM experience (mean numbers). ............................................................. 40

Figure 15: Standard deviations of the ratings of BIM’s effect on different aspects of

designing by BIM users. ............................................................................................. 41

Figure 16: Open-ended question about the features of BIM – answers categorised and

arranged in a stacked bar chart. .................................................................................. 42

Figure 17: Open-ended question about the problems of designing – answers

categorised and arranged in a stacked bar chart. ........................................................ 43

Figure 18: The occurrence of topics and sentiments in workshop discussion: Case A.

.................................................................................................................................... 49

Figure 19: The occurrence of topics and sentiments in workshop discussion: Case B.

.................................................................................................................................... 58

VII

Figure 20: The occurrence of topics and sentiments on workshops: Cases X-Y. ...... 68

Figure 21: The design tasks as shown in project schedules ....................................... 76

Figure 22: Open-ended question about the features of BIM – answers categorised and

arranged in a stacked bar chart. .................................................................................. 79

Figure 23: BIM process in the case projects, stage 1 of BIM implementation. ......... 84

Figure 24: Potential stage 2 of BIM implementation. ................................................ 89

Figure 25: Potential stage 3 of BIM implementation ................................................. 90

Figure 26: Focus areas for design companies in infrastructure sector ....................... 93

Figure 27: Integration of various types of data on the project timeline. .................. 100

List of tables

Table 1: The possible directions for process improvement in design projects, based on

the reviewed literature ................................................................................................ 27

Table 2: Case projects ................................................................................................ 32

Table 3: Analysing the workshop recordings (fragment) ........................................... 36

Table 4: Experience of workshops participants, reported in questionnaire. .............. 38

VIII

Terms and abbreviations

BIM Building Information Modelling

AEC Architecture, Engineering and Construction –

industry sector and activity

Infrastructure (projects) In this work – transport infrastructure,

particularly land transport

Lean Philosophy of improvement in productive

activities, originally developed in Japan

CE Concurrent Engineering

Agile A set of methodologies for managing software

and product development projects

Collaboration model A common model that integrates all discipline-

specific data

Model coordinator (integrator) A BIM-related role responsible for modelling

supervision and integration of discipline data

Interoperability Depending on the context: organisational

interoperability and software interoperability

Initial design, conceptual

design

The first, undetailed stage of design in

construction projects

General design The second, more detailed stage of design

development in construction projects

Construction-level design The last, most detailed stage of design

development

1

1. Introduction

1.1. Background of the study

The rapid development of information and communication technology (ICT) since the

second half of the last century has resulted in new challenges for businesses. So far,

ICT has offered enormous advantages and ultimately changed the way businesses

work; however, at the dawn of the ICT revolution these advantages were not as

apparent as they seem today. Even nowadays, given the high number of information

and communication solutions entering the market each year, it may sometimes become

a difficult task for a company to decide if it should invest in acquiring a particular ICT

product or not. Such investment does not exclusively relate to the offering itself, but

also to the development of the necessary skills and process change in the company.

Since the early years of ICT adoption in industries, a body of research has formed to

address the issues of exploring and evaluating the benefits of technology, and assessing

the associated return on investment (ROI) (Mahmood & Mann, 1993; Weill, 1992;

Davern & Kauffmann, 2000). To a large extent, the necessity of such studies was

dictated by the fact that many businesses were hesitating in technology adoption,

demanding a strong ‘proof’ of the benefits of technology. When the potential impact

of technological advances on business activities and even the potential use of many

technologies is unclear, making investment decisions is difficult. Even nowadays, the

implementation of complex technologies and processes in growing companies can

cause hesitation.

In today’s architecture, engineering and construction (AEC), the development of new

information technology is largely related to Building Information Modelling (BIM), a

set of technologies and processes for integrating and managing building information

throughout the building lifecycle. The rapid technology adoption in the sector means

that the companies have to face the problem of justifying their investments.

The concept of BIM includes a 3D modelling process, a related technology and also

an information management approach. Unlike CAD solutions, BIM allows faster and

more accurate modelling of structural elements and systems by using parametric

object-based modelling technology, and supports integration of data in a single

collaborative 3D model that can be used in construction and maintenance phases of

life cycle (Eastman et al., 2008).

A commonly acknowledged benefit of BIM is detection of clashes between building

elements that typically occur due to design inconsistencies and can result in major

rework and cost in construction phase. Other benefits include modelling of

construction site and coordination of construction work, accurate quantity take-off and

cost estimation, 3D visualisation of buildings and support for collaboration between

2

different technical areas from design to operation phase (Eastman et al., 2008). The

ultimate (and yet not completely achieved) goal of BIM is seamless integration of all

building information to create a virtual building environment, so that the variety of

building information is available for all users on all stages of building life cycle

(Succar, 2009).

A number of studies call for quantifying the benefits of BIM and ROI related to BIM

implementation, and development of metrics for measuring BIM performance, which

are essential for setting organisational goals and evaluating the current state of BIM

process (Succar, Sher and Williams, 2012; Coates et al., 2010; Mom and Hsieh, 2012;

Barlish and Sullivan, 2012).

Barlish and Sullivan (2012) after analysing over 600 literature sources have identified

only 21 studies that mentioned benefits of BIM explicitly. Only 4 studies contained

empirical evidence derived from case study data. None of the studies provided

methodology for calculating the return on investment (ROI) in BIM or comparing BIM

and non-BIM projects. One of the likely reasons for lack of quantitative studies in this

area is the fact that the companies most often do not collect extensive data on project

performance that could allow developing reliable, if any, quantitative measurements.

In addition, a large share of possible benefits of BIM relate to the qualitative rather

than the quantitative aspects of companies’ performance. It can be difficult, if not

impossible, to directly measure the impact of BIM on such abstract performance

criteria as the quality of design solutions or the degree of mutual understanding

between project stakeholders. Finally, reliable comparison between projects is difficult

because each project is unique in terms of objectives, personnel, timescale, customer

requirements, engineering difficulty and external environment. The lack of

quantitative performance measurements is most noticeable in design and engineering

activities because the performance criteria in these creative activities are difficult to

formulate and quantify, and because the value of intangible outcome cannot be directly

connected to resources or costs.

These difficulties in measurement relate to the inherent characteristics of services.

Both BIM software solutions and architecture and engineering consultancy are

services. The service characteristics include the intangibility of outcome and variations

in process and resources due to customer involvement (Moeller, 2010).

To overcome the problems with direct measurement, the performance of services is

frequently assessed indirectly through the perception of the clients, using various

qualitative and quantitative approaches. The idea of measuring customer satisfaction

to assess product quality is common in marketing and can be applied for both services

and goods. There is a body of research that develops methods for measuring customer

3

satisfaction and perception of quality specifically for service industries, e.g. the

SERVQUAL model (Parasuraman et al., 1985).

The approach of evaluating the perception rather than measuring the financial or

process parameters is relevant to services within the AEC sector. Such exploration of

perceptions can be focused on the qualitative benefits of ICT and process change,

which may be hard to quantify. The study of perceptions can also provide the initial

knowledge for developing direct, quantitative performance measurement in the future.

1.2. Research scope and objectives

The current research aims at exploring the perception of benefits of BIM and the

process improvement in the design projects in transport infrastructure sector. Since the

new technology typically requires companies to change their processes and work

practices, the research examines the experts’ vision of process change and

organisational change and its relation to BIM implementation. It discovers challenges

faced on the design phase of infrastructure projects, and examines the capability of

BIM to become the enabler of collaborative design. The practical objective of the study

is to provide recommendations for different industry stakeholders and to suggest

directions for the future research.

According to the initial plan, the main objective of the research was to apply

quantitative methods to understand the impact of BIM on project performance.

However, the initial phase of the research revealed that, given the available data,

achieving this goal was not possible. Neither of the available companies did collect

sufficient data in their projects. In addition, the timeframe of the research did not

permit to follow the entire projects, applying the right methodology from the beginning

of the projects until the end, as required to make comprehensive measurements. As a

result, the research objectives were reformulated. However, some recommendations

on how to carry out the measurements in future are formulated in chapter 5 based on

the experience gained during the study.

The research objectives can be summarised as following:

Understand the factors that can make BIM beneficial for design

Examine the current state of design practice and BIM implementation in

different infrastructure projects, identify problems and opportunities

Evaluate the perception of BIM and design process by the project teams

Utilise the theoretical and practical knowledge to provide a framework for

future BIM and design process development

Provide recommendations for future research

4

The formulation of the research objectives leads to the following research questions.

The first research question aims at discovering the theoretical concepts that could

explain the factors of successful design. The question is mainly addressed in the

literature review chapter.

RQ1. What process-related and organisational factors determine the efficiency

and effectiveness of collaborative design in construction projects?

The second research question explores the potential role of technology in achieving

better performance in design, based on the factors explored in the previous question.

This question is addressed in both the literature study and the empirical study.

RQ2. What is the potential role of BIM and other technologies in maximising the

efficiency and effectiveness of collaborative design in construction projects?

The third question evaluates the current state of process and technology in the industry

based on the knowledge obtained when answering the questions 1 and 2. The current

state of industry is examined in the empirical study. The third question relates to both

the objective indicators of the industry’s state and the perception of professionals,

which plays a major role in this research.

RQ3. What is the current state of design practice and BIM implementation in the

infrastructure sector in Finland? How do the professionals perceive it?

The last research question concerns the theoretical and practical implications of the

results of the study, which are discussed in the last chapter.

RQ4. What actions can be taken by the design companies in infrastructure sector

regarding the technological and process development in order to increase

performance in design activities?

The research framework based on the research objectives and research questions is

visualised on figure 1.

1.3. Research motivation

The choice of design phase of infrastructure lifecycle as the focus of this study is

related to the low number of existing studies that could describe the use of BIM in

actual design projects and could explore a connection between the potential benefits

of BIM and design process change. The benefits of BIM on construction phase are

generally explored better than benefits on design phase.

5

Figure 1: The research framework

At the same time, the design phase of lifecycle has a critical influence on building’s

performance and has the largest impact on lifecycle costs when compared to other

phases (figure 2). In the end, it is architects and engineers who initially deliver BIM,

and therefore the design companies must be able to feel the benefits of BIM.

Figure 2: Influence of overall project cost over the project lifecycle. Adopted from Eastman et

al., 2008: 99

The research is focused on infrastructure projects. Discussions with industry experts

held before the research have suggested that the adoption of BIM in the infrastructure

sector worldwide is behind the adoption in building sector. The literature review

Ability to influence costs 100%

0% 0%

100%

Lifecycle

Lev

el o

f in

fluen

ce o

n c

ost

Conceptual

planning/ feasibility

studies

Design and engineering

Procurement and construction

Start-up

Operation and

maintenance

Construction cost

6

undertaken as a part of this work has revealed that the implementation of BIM in

infrastructure construction projects is not sufficiently studied.

The role of design process and focus on process in the study require a separate mention.

The particular assumption about BIM to be investigated in this study is that it can

facilitate the efficiency of design process through supporting collaboration and

communication in design teams, particularly between different technical areas. The

expectation is that BIM is most beneficial when implemented together with the right

process, creating a sustainable synergy. The studies and expert opinions emphasise the

connection between BIM implementation and process improvement.

A view of organisation as a socio-technical system commonly found in literature on

Lean, often includes three dimensions: people, process and technology. These

dimensions are outlined in a book on Toyota Product Development System by Morgan

and Licker (2006), and are also present in BIM-related articles (e.g. Arayici et al.,

2011). The study by Succar (2009) describes the implementation of BIM from the

point of view of three nodes: technology, policy and process. Figure 3 combines them

into four components of organisational performance: people, process, technology and

policy.

Figure 3: The four components of organisation’s performance: people, process, technology

and policy

Although new technology can cause process change, the technology adoption should

not be viewed as a primary or a more important concept compared to process

improvement. It is important to note that this study, while exploring the relation

between BIM and design process, will consider process improvement an independent

conceptual entity. A variety of practices have been developed in AEC and other

industry domains to maximise the effectiveness of designing in the existing

circumstances and business environment. BIM and other technology can enhance or

disrupt these practices, impact the efficiency of the design activity in a positive or a

7

negative way. The technological changes are calling for major shifts in conception of

how design work should be carried out.

The research aims at exploring what practices support the efficiency and quality of

design in a best way in the current circumstances and if the technology (not necessarily

or only BIM) can enhance the positive impact of these practices. For example, the

research investigates the potential application of data management tools and computer-

aided project management and control in infrastructure design firms. It also

investigates whether substantial change in design process would be needed in future

to achieve the synergy between technology and process.

The design process in this research will be examined from the perspective of Lean

philosophy. This set of approaches was initially developed by Japanese car

manufacturers, but was later applied to other industries and activities. Lean philosophy

aims at identifying and eliminating different types of waste that occur in business

activities, in order to maximise value delivered to the customer. Any action, operation

or time period where no value is added to the product constitutes waste and should be

excluded from the process if possible.

The value and waste concepts are intuitively simple and close to how people

understand efficiency. Lean methodology, nevertheless, is very comprehensive and

includes a variety of practices applicable to different kinds of activities in different

industries. The principles of Lean and their application to design process will be

discussed in detail in chapter 3.

1.4. Research planning and the literature study

The research goals were addressed in two stages: (1.) literature study and (2.) empirical

study. The literature study provided the basis for formulating the research questions

and planning the empirical research. It was divided into two phases. The first phase

included the initial exploration of the research field and the formulation of research

objectives and questions. These questions and objectives were refined later during the

research process. The majority of this phase’s outcomes have been already presented

in the first chapter.

The second phase of literature review addressed the research questions in detail, and

was particularly focused on research questions 1 and 3:

RQ1. What process-related and organisational factors determine the efficiency

and effectiveness of collaborative design in construction projects?

RQ2. What is the potential role of BIM and other technologies in maximising the

efficiency and effectiveness of collaborative design in construction projects?

8

These general research questions were broadened during the literature review when

the variety of topics addressing these questions was discovered within the literature,

because none of the studies could answer the questions comprehensively.

Due to the lack of literature within the AEC field that could address the research

questions, the review included a number of studies that could create a theoretical basis

for process improvement in the sector but were not related to the AEC directly. This

included the fundamental research on design ontology and cognition, innovation and

knowledge management, communication, collaboration, lean and process

improvement, information management, software engineering and product

development. This was beneficial for the study because it allowed bringing the notions

from different fields together and benchmarking the concepts.

This conceptual approach and the broad scope of research topics required a narrative

and exploratory way of conducting the review. At the beginning of the study a list of

questions and topics was prepared to define the directions of the search. Different

combinations of search words related to those topics were then used to find the relevant

scientific articles. The relevance of the articles was assessed according to the following

criteria:

Relevance to the search topic;

Scope of the study: industry sector, location.

Source of publication: the review considered only the articles published in

journals and conference proceedings, published dissertations and books;

Number of citations;

Year of publication.

After the new topics emerged, the questions were refined and the search process

reiterated. The relevant references cited in the literature were also examined. The

search was conducted through Google Scholar and complemented by Aalto University

Library and scientific publication databases, such as ScienceDirect. In addition,

notable books and articles known prior to the study were reviewed.

The literature review is presented in chapter 2. Section 2.1 of the chapter provides

general overview of BIM and AEC business environment. The later sections (2.2 –

2.4) describe the ontology and conceptual characteristics of collaborative design from

managerial point of view, and explore the potential of applying lean approach to design

practice. Section 2.5 discusses the potential benefits of ICT and particularly BIM for

design stage of construction projects. The last section (2.6) is a brief summary of the

conclusions and hypotheses formulated as the result of literature review, which have

shaped the empirical study.

9

Chapter 3 describes the methodology of the study. Chapter 4 describes the case

projects and presents the results of the empirical study. Chapter 5 includes the

discussion of results, theoretical and practical implications. The sequence of the

research process is illustrated on Figure 4.

Figure 4: The research process

2. Theoretical foundations

2.1. Overview of Building Information Modelling

According to Khanzode (2012), today’s construction industry is characterised by high

complexity, especially in highly technical facilities such as hospitals and data centres.

This has led to high specialization and vast array of subcontractors and consultants in

projects. The parties and disciplines include key trades (such as mechanical,

engineering and plumbing - MEP) and auxiliary trades (glass and glazing, elevators,

landscaping, pneumatic tubes etc.). According to statistics provided by Eastman et al.

(2008: 9), sixty-five percent of construction firms consist of less than five people. At

the same time, the average number of companies participating in a construction project

is 420, and the average number of individuals involved is 850. The coordination of

work of different parties on different stages of a project and the corresponding

information exchange are complex and massive processes. The average number of

pages of documents generated in a project is 56 000. (Eastman et al., 2008: 2-3)

According to numerous reports and studies, the productivity of construction sector is

significantly behind other non-farm industries. The sector suffers from low level of

Initial research

planning,

familiarising with

the problems

Literature review.

Finding case projects

for empirical study

Reformulating the

research objectives.

Selecting research

methods.

Conducting

interviews with

project managers.

Planning workshops.

Preparing the

questionnaires and

workshop questions.

Conducting first

workshops and the

survey. Intermediate

data analysis.

Preparing for and

conducting the

second series of

workshops.

Analysing the data.

Additional readings

and evaluation of the

findings.

Drawing conclusions

and making

suggestions.

Preparing reports.

10

automation, low labour productivity and lack of productivity growth (Eastman et al.,

2008: 8-10; Bergly, 2001; Aapaoja, 2014: 17).

Aapaoja (2014: 17) synthesises the relevant articles and suggests the following list of

problems of the construction sector:

the costs are finally accumulating to the customer,

major deficiencies in several different fields of know-how (e.g., design,

quality, lead time, and predictability of work),

differences (e.g., process, methods, viewpoints, opinions) between project

organization and both internal and external stakeholders are not taken into

account in the best possible way,

functional decentralization and inefficiency,

inability to collaborate,

lack of processes to transfer and develop the knowledge and competencies, and

harsh sub-optimization during the project

The approach to managing construction information traditional in AEC is

characterised by inefficiency. During the design phase of construction projects, each

design discipline creates its own set of engineering documents to be provided along

with typically less detailed general documents. This documentation is passed forward

to the subsequent designers and then to a building contractor. In building construction,

there are also subcontractors who create detailed construction-level documentation

within their own technical areas. (Eastman et al., 2008: 4-6)

Because of high number of participants and differences in standards, the construction-

level documentation is prone to mistakes and inconsistencies. These mistakes and

inconsistencies, if not identified prior to construction, can make construction

companies delay their work and even rebuild some of the structures, which can

significantly increase construction costs. In addition to errors, designs can require

changes during construction due to unanticipated site conditions, changes in available

materials and technologies, and new customer requirements. Clarification of design

documents requires construction contractor and subcontractors to send requests for

information (RFI) to designers, which can be very time-consuming. Incurred losses

often result in legal disputes between project participants. (Eastman et al., 2008: 4-6)

The case studies conducted by Fulford and Standing (2014) suggests that high

fragmentation in the industry, disparate project management processes and non-

standardised information contribute to low productivity. As they summarise in the

article,“(1.) the construction industry lacks the ‘strength’ of relationships necessary to

create a network of organisations that trust and have shared values; (2.) design

processes should include both value engineering and lifecycle costing; (3.) procedures

11

and information need to be standardised; (4.) there should be more emphasis on value

adding project management activities.”

Building information modelling (BIM) has emerged from 2D and 3D CAD solutions

in response to numerous problems and requirements specific to construction that were

not addressed by CAD and paper-based information process (meaning both physical

paper and its digital equivalents). Although the wide adoption of BIM has started

within the last decade, the term has been known for almost 20 years, and the core

principles of BIM were outlined as early as in 1975. (Eastman et al., 2008: xi)

BIM is a process of creating and managing the digital representations of building’s

spaces, elements and systems, including their spatial, functional, temporal and

economic characteristics. Like 3D CAD, it represents the spatial characteristics of

building’s elements and surroundings in three-dimensional environment, making it

possible to visualise the building in detail. (Eastman et al., 2008: 13)

The core technological difference of parametric object-based BIM from CAD is that

BIM typically deals with meaningful objects, parameters and relationships between

objects rather than pure geometric shapes and surfaces. Drawing the whole building

with all its details manually in CAD is a difficult or sometimes impossible task,

especially when frequent changes are required in drawings. BIM systems compute the

models according to object rules and parameters set by a user, allowing automation of

changes and automated checking for consistencies. (Eastman et al., 2008: 14)

The conceptually important aspect of BIM approach is that it aims at combining and

centralising all construction drawings and data throughout the whole life cycle of a

building. BIM can integrate all 3D models of different building systems and elements

in a single virtual environment, called collaboration model or build model, and

integrate other construction and engineering data in connection with different 3D

objects in that model, including materials, cost and various technical details. Initial

data that describes situation before the project, for example existing structures, terrain

and surroundings, is also integrated in the model.

This approach gives a number of benefits to the project. The technology allows

visualising the building and providing all companies involved in a project and all

individual members of a project team with comprehensive and consistent data. To

decision-makers, BIM gives additional support in formulating the requirements and

choosing between alternative solutions. To designers, it can work as a collaboration

tool helping to coordinate the work of different technical disciplines. It allows

detection of errors and clashes in designs (e.g. clashes between pipelines and building

foundations) early in the design phase. (Eastman et al., 2008: 16-21)

12

BIM can also facilitate comprehensive engineering calculations and construction cost

calculations during design, as early as on conceptual design phase. Such systems can

combine spatial 3D data and other construction and engineering data from different

technical domains. (Eastman et al., 2008: 18; Sattineni & Bradford, 2011)

The shared 3D model helps the building contractor and subcontractors to coordinate

their work on construction site and eliminate possible conflicts (Eastman et al., 2008:

19-20). Advanced BIM tools can model the temporal (4D) and economic (5D)

characteristics of a building and building process, to create a visualised simulation of

construction work with the corresponding schedule, material supply and cash flow

calculations connected to the project’s timescale. This is used for planning the

construction work, including its financial and logistical aspects (Eastman et al., 2008:

18; Kang, Anderson and Clayton, 2007). After construction is finished, the models can

facilitate efficient operation and maintenance (Eastman et al., 2008: 20-21). BIM is an

intelligent approach to creation and handling of building information throughout the

whole lifecycle of the building.

M. A. Mortenson Company’s definition of BIM technology, cited in BIM Handbook

(Eastman et al., 2008: 13), describes BIM as an “intelligent simulation of architecture”,

and gives the following definitive characteristics of such simulation:

Digital

Spatial (3D)

Measurable (quantifiable, dimension-able, and query-able)

Comprehensible (encapsulating and communicating design intent, building

performance, constructability, and include sequential and financial aspects of

means and methods)

Accessible (to the entire AEC/owner team through an interoperable and

intuitive interface), and

Durable (usable through all phases of a facility’s life).

Although this initial description of BIM is short and general, it provides some view on

the potential impact of BIM on construction industry. The BIM Handbook states no

current implementations of BIM meet all of the listed criteria (Eastman et al., 2008:

14).

The ultimate goal of BIM innovation, as it is often seen, is integration of all participants

in a shared collaborative project delivery process, based on BIM. Succar (2009)

divides BIM implementation in three capability stages, which are defined by minimum

requirements that organisations have to meet in order to qualify for each stage:

13

1. Object-based modelling – organisation deploys an object-based modelling

tool. On this stage BIM is only used within one organisation. An organisation

produces models to enhance the quality of its work but the models are not

transferred to other parties.

2. Model-based collaboration – organisation is involved in multidisciplinary

model-based collaboration. On this stage several parties in the project use BIM

and they are able to transfer data to each other.

3. Network integration – organisation shares object-based models with at least

two other organisations. On this stage several parties share integrated models

throughout the project, and their software is able to use the models and

interchange data. This can be achieved using model server technology.

After network integration is achieved, the industry continues to develop to post-BIM

stage, with ever-evolving goal of employing virtually integrated Design, Construction

and Operation (Succar, 2009).

The literature provides the connection between the benefits of BIM and the use of

relational project delivery arrangements. Traditional contracting arrangements, and

particularly design-bid-build (DBB), are used in the largest share of today’s

construction projects and in up to 90% of projects in public sector (Eastman et al.,

2008: 3-5), since some form of tender is typically required for public contracts by state

legislations. The majority of projects in the infrastructure sector are public, and

therefore this procurement method dominates the sector. In DBB, the design tasks,

consultancy, construction work, maintenance and other tasks are contracted separately

to individual parties not directly connected with each other by contractual

relationships. This procurement method is usually based on the lowest bid. (Eastman

et al., 2008: 4-7; Aapaoja, 2014: 17).

The traditional DBB approach suffers from numerous flaws. This contracting method

does not allow involving subsequent design stakeholders in the early phase of the

project, limits collaboration, information exchange, knowledge transfer and innovation

(Eastman et al., 2008: 4-6; Aapaoja et al., 2013; Aapaoja, 2014; Matthews and Howell,

2005). In addition, the sequential work process favours point-based design – instead

of optimising the overall value of the project, local optimisation occurs in different

technical areas (Aapaoja, 2014: 17; Matthews and Howell, 2005). There is a tendency

to rush over conceptual planning phase to detailed design and to construction in order

to limit the costs, without proper understanding of project requirements and objectives,

which can undermine the project’s value. (Aapaoja, 2014: 17-18)

An implication of the traditional approach for information exchange is inefficient

information flow caused by lack of interoperability. This is a large concern with regard

to BIM implementation. A large variety of specialised modelling and engineering

software exists that are not always interoperable with each other, which makes BIM

14

hard to extend over the building lifecycle and among the project participants. The study

by the US National Institute of Standards and Technology (NIST) showed that

inefficient interoperability accounted for a significant increase in construction costs

(Gallaher et al., 2004, cited in Eastman et al., 2008). The key reasons for the cost

increase as suggested by the study were (1.) redundant computer systems and

inefficient business process, (2.) manual data re-entries and requests for information

(RFI), and (3.) delay costs associated with idle employees and resources.

The adoption of Relational Project Delivery Arrangements (RPDA), in which all

project participants are unified by a common contract, can become an important

facilitator of productivity improvement in construction (Aapaoja, 2014; Lædre and

Haugen, 2001). There is evidence that such arrangements result in cost savings in

construction projects (Lædre and Haugen, 2001). Early stakeholder involvement and

integration enhances value creation in projects. (Aapaoja et al., 2013; Aapaoja, 2014).

Integrated project delivery arrangements that are based on collaboration and shared

risks allow the achievement of BIM-based network integration (Succar, 2009).

The innovation in the AEC sector is fostered by the rising complexity of the buildings

and increasing requirements, especially those related to environmental impact,

maintainability and energy efficiency (Khanzode, 2012). The development of ICT and

process improvement in the AEC can create conditions for productivity improvement

that is so much appreciated by the companies, clients and public. The implementation

of BIM and other intelligent technologies, effective procurement methods and

collaboration practices is led by various industrial associations and public

organisations, such as educational and research institutions, territorial administrations

and public authorities.

2.2. Ontology of collaborative design

The design and engineering in most of the industries is a collaborative process that

involves a number of individuals with expertise in different technical fields. The

design phase of construction projects can be divided into several stages (Eastman et

al., 2008: 151-152):

1. Pre-design. The construction projects typically start with the pre-design, on

which the need for construction is assessed and initial building requirements

are formulated. In infrastructure projects this phase involves customer, public

stakeholders and consultants, and is typically merged with the second design

stage.

2. Conceptual design. Initial, conceptual and schematic design are

interchangeably used names of this design phase aimed at formation of design

solutions, definition of shapes and spaces, materials and building systems. In

infrastructure projects, feasibility studies are usually conducted

15

simultaneously with this stage to assess the economic feasibility of undertaking

the project.

3. General design. On this stage (also called design development) the building

plans, major building systems and materials are taken to a higher level of detail.

The essence of the design solutions and the construction costs are defined on

this stage.

4. Construction-level design. On this design stage, the final set of highly-

detailed documents is generated for use in construction. It includes finished

detailed plans of all building elements, specifications of systems and materials,

plans of site work, and acceptance criteria for building systems. (Eastman et

al., 2008: 151-152)

In order to understand the factors that determine the efficiency and effectiveness of

collaborative design a closer view on the nature of design and collaboration is needed.

Design is a human activity that strives to improve the environment through creation of

artefacts. This is done by formulating functions to be achieved and creating design that

is made to perform these functions. Design is a purposeful activity that transforms the

requirements (an envisioned function) into a design description of an artefact, and this

description can be used to produce the artefact. Design description is the main outcome

of design (Gero, 1990).

According to the Function-Behaviour-Structure (FBS) framework developed by Gero

(1990), design process involves the following sub-processes:

Formulation. Designer formulates how the structure must behave in order to

perform the desired function.

Synthesis. Designer generates the structure.

Evaluation. Designer compares the actual behaviour of the structure with

expected behaviour.

Reformulation. If the actual behaviour is different from the expected

behaviour, designer reformulates either the structure (type 1), or the expected

behaviour (type 2), or the function (type 3).

Documentation. Design description of the structure is created.

The design is performed in two contexts: the one within which designer operates and

the one produced by developing design itself (Gero, 1990). The FBS framework is a

notable framework that has been applied in design-related studies in different areas.

The concept of situatedness developed in Gero’s and Kannengeisser’s (2004)

following work on design ontology relates to a notion that designer’s decisions to

change the environment are affected by the current state of the environment (e.g. by

existing design). “This means that the designers’ concepts may change according to

16

what they are “seeing”, which itself is a function of what they have done” (Gero and

Kannengeisser, 2004).

These concepts are important because they provide a generalised view on the design

process as an iterative process in which change, reformulation and re-documenting are

inherent. The design environment created by the designer affects how he sees the

design later and what actions he takes next. In collaborative design, a designer is not

only affected by his own actions, but also by those of other designers.

The development of design by a group of individuals requires shared vision of function

and structure that is achieved through communication. The designers have to

coordinate their individual inputs in design to be able to maintain this shared vision

and to fit the elements of design together. The meaning of collaboration is that through

the combination of specialised knowledge of designers the result is achieved that could

not be achieved by any of the participants working alone. (Kvan, 2000)

The integration of knowledge is the goal of collaboration. Kleinsmann (2006, cited in

Kleinsmann, 2008: 370-371) defines collaborative design (collaborative design) as the

“process in which actors from different disciplines share their knowledge about both

the design process and the design content. They do that in order to create shared

understanding on both aspects, to be able to integrate and explore their knowledge

and to achieve the larger common objective: the new product to be designed.”

The concept of transactive memory, developed by Wegner (1995), provides a process

view on collaboration. Transactive memory occurs when an individual serves as an

external storage of information for another individual. The individuals rely on common

memory in order to retrieve information that each of them does not know, from other

individuals. Transactive memory allows combining the memory of all the group

members. (Wegner, 1995)

The notion of memory in the studies on information systems refers not specifically to

past events, as intuitive understanding might suggest, but to information contained on

a particular storage (Wegner, 1995), in the current case – an individual’s mind. Thus,

the transactive memory includes information that existed before the design has started,

including the professional knowledge and experience of individuals. The notion of

memory in this case is equal to knowledge.

To retrieve information from the transactive memory, one must find its location; in

other words, know who could provide such information. But each person does not

know at every point of time what other people know. Each person is familiar with the

knowledge-holding system from his own perspective, but the system is larger and more

complex than that of any of the individuals (Wegner, 1995). Therefore, the knowledge

17

of the state of the knowledge-holding system is required at each time to ensure

effective information exchange (Kleinsmann, 2008).

Collaborating designers develop shared understanding of the design, but also of the

knowledge-holding system. “Shared understanding is a similarity in the individual

perceptions of actors about either how the design content is conceptualized (content)

or how the transactive memory system works (process)” (Kleinsmann, 2008).

The concepts of situatedness, transactive memory and shared understanding provide a

basis for viewing collaborative design as a temporal process, in which the changes to

the environment made by designers are affected by the sum of past actions that have

formed the existing state of the environment.

In addition, the design is affected by the state of environment and the knowledge

existing prior to the project. For example, a number of challenges in communication

stem from differences among actors that exist initially before the design project starts.

Those differences relate to initial knowledge and skills, vocabulary, professional

values and goals (Kleinsmann, 2010; Pei, Campbell and Evans, 2009).

Following the point of view that communication is an autopoietic (self-organised, self-

steered) system that is characterised by high level of uncertainty, Maier, Eckert and

Clarkson (2005) argue that the possibility of any control over communication is

limited. “Influence, however, is possible if one manages to understand and connect to

the internal logic of a system, i.e. the components and its connections and its specific

rules by which it operates. The preferred solution seems to be to raise awareness and

engage the designers in an ongoing learning process.” (Maier, Eckert and Clarkson,

2005).

The notion of ‘internal system’ and its ‘logic’ in this statement shows similarity to the

concepts of transactive memory and shared understanding discussed above. The

concepts of transactive memory, shared understanding and communication as a social

system suggest that the design is affected by designers’ knowledge about the team

(knowledge-holding system) and teamwork process, and not only by the understanding

of design object.

The framework presented on figure 5 illustrates the system of professional

collaborative design based on the review of the literature on collaborative design

presented in this section.

The framework depicts two specialised designers working together. The visible

outcomes of their work include the design description and the information documented

during the design that is not attributed to the design description (e.g. documented

requirements, communication, solutions and other documented knowledge).

18

As illustrated in the framework, the design develops as a result of a cognitive process,

in which the knowledge of many designers is combined. Here, memory incorporates

all information possessed by designers at every point of time, and generally consists

of the initial knowledge that occurred before the design started, and new knowledge.

Designers develop both the knowledge of object being designed and the designing

system (e.g. knowledge structure in the team). The combination of designers’ memory

forms the transactive memory. The framework is a proposal, meant to visualise the

collaborative design system based on the reviewed articles. It has not been practically

applied or tested in this study.

Figure 5: The conceptual framework of collaborative design system in professional domains

2.3. Application of Lean principles to collaborative design

‘Lean’ in a set of concepts, principles and methods aimed at increasing the efficiency

and effectiveness in productive activities. It was developed in post-war Japanese

automotive industry, which was rapidly developing and was later able to succeed on

worldwide market competing with local counterparts, for instance, with US

manufacturers on the US market.

The first and most fundamental ideas of lean manufacturing and philosophy were

derived from the principles of Toyota production system (TPS) (Ohno, 1998). The lean

approach focuses on delivering value to the customer and eliminating overburden

(‘muri’), inconsistency (‘mura’), and waste (‘muda’) out of production system.

Designer B Designer A

Design

description

Cognitive

process

Cognitive

process

Communication

gn activity

External storage: documented requirements,

communication, solutions, other knowledge.

Knowledge of

designed object

Knowledge of

designing

system

Knowledge of

designed object

Knowledge of

designing

system

Professional background Professional background

Transactive

memory

19

Waste is any activity that absorbs resources but does not contribute to value-creation:

processing steps that are not needed, unnecessary stocks of materials, transportation,

waiting for the upstream activity to be finished (Womack, 2003). Toyota production

system outlines 7 basic types of waste (Ohno, 1998), which are typically found in

manufacturing systems:

1. Waste of over-production (too much or too early production at cost of

additional resources)

2. Waste of time on hand (waiting, inactivity of people and tools)

3. Waste of transportation (unnecessary movements of products and materials)

4. Waste of processing (more work or higher quality than required)

5. Waste of stock at hand (products and materials waiting to be processed)

6. Waste of movement (unnecessary movements and actions of people)

7. Waste of making defective products

The implementation of Lean starts from identifying what constitutes value in

production system. In case of design, a creative activity with immaterial product, it is

hard to define value and the sources of value, and to connect them to resources.

Aapaoja (2014: 33-34), based on a number of studies, describes value as a relationship

between benefits and sacrifices achieved through customer relationships. From the

customer’s perspective, value relates to quality – the degree to which his requirements

are satisfied, and to which the design fits the purpose.

However, quality is not the only factor of design value. One can argue that more

original and innovative solutions constitute more value than generic designs and

require more resources and expertise to produce. Eastman et al. (2008: 154) describes

the concept of information development, and suggests that different construction

projects can be at different levels of information development, from standardised to

highly innovative. The studies by Brown and Chandrasekaran (1985) and Brown

(1996) identify three categories of design solutions in terms of creativeness,

determined by the number of possible solutions and combination of variables in

design: routine, innovative and creative design. Coyne et al. (1987) argue that the

creativeness of design is defined by entropy (how much knowledge is available prior

to design work), efficiency (time restrictions) and richness (number and variety of

produced alternatives).

These concepts and the concepts found in the studies on collaborative design reviewed

in section 2.4 view design largely as an information process. Indeed, the inputs and

outputs of design are information (Kiviniemi, 2005: 26). Eppinger (2001) in his paper

on management in product development projects mentions that the manager who wants

to optimise the workflow in design-related projects needs to concentrate on the

information flow and to understand the information needs of every task. For example,

20

when a product designer has to work with incomplete information from the previous

task this may lead to rework (Eppinger, 2001).

It is logical to apply these statements to the AEC sector and to suggest that the

opportunities for process development in design activities can be found in the domain

of information management. This hypothesis is significant from the perspective of IT

and BIM in particular.

Studies on information systems and information management contain parallels to the

problems found in design process. For example, Hicks (2007) in his overview of lean

approach to information management suggests that the notion of waste in information

management relates to additional actions or inactivity that arise as a consequence of

not providing appropriate and correct information to consumer when it is required.

A good example of such inactivity in construction sector is the process of sending a

request for information (RFI) to another project stakeholder and waiting for the

response (Eastman et al., 2008: 5-6). In collaborative design, the unavailability of

information can be caused by the lack of input data, and can force designers to wait or

work with incomplete (possibly wrong) information. Unavailability of information,

therefore, can cause waste of time. The goals of information management are

“optimising the creation, representation, organisation, maintenance, visualisation,

reuse, sharing, communication and disposal of information”. (Hicks, 2007)

The studies on design stage of construction projects and concurrent engineering

reviewed above suggest that design changes that occur because of mistakes or due to

normal reformulation process, can cause re-work, which requires designers to spend

additional time and undermines some of their previous work. While changes are

inherent in a normal design process, their impact can be minimised if the need for

change is recognised early enough in design process (see the next section). This is

generally addressed by reducing the time between interrelated tasks and this way

making the iterations shorter (Eppinger, 2001).

An important point that is not presented in the reviewed literature but needs to be

mentioned, is that industrial approaches should be applied to design activity with

caution. Design is a socio-cognitive activity, and not a controllable industrial process.

Some activities in design that seemingly constitute waste (create no value) might be

an important step in the cognitive process of design that will later result in development

of design solutions. Even a seemingly wasteful activity, such as re-work, may result in

a new understanding or idea that can contribute to the final outcome.

The search for waste in projects should be therefore focused on more tangible sources

of waste, for example, unnecessary waiting. While the socio-cognitive element of

collaborative design is self-regulating and hardly controllable, it is possible to control

21

the structural, mechanistic element including the way design information is handled,

so that support is provided for the socio-cognitive process in terms of instruments and

organisation.

2.4. The characteristics of collaborative design in project management

A number of conceptual problems result from the nature of collaborative design

process. The early attempts to describe the design process have viewed designing as a

sequential activity, in which the design evolved in a series of steps from early

conceptualisation to testing and release. The actual design projects at that time were

planned according to this principle.

In the software industry, however, this manufacturing-like method of design proved to

have significant flaws. In the middle of the last century, the early industry was

characterised by high rate of delayed or failed projects (Larman, 2004: 101). It was

typical that the project teams created hundreds of pages of paper-based documentation

in order to manage information, but such detailed documents were incomprehensible

(Larman, 2004: 63-108).

The Waterfall model, first formulated by Royce (1970) in his well-known paper on

development of large software systems, was presented in an article that criticised the

linear software development. The model became a classic illustration of a linear design

process applicable in other domains of design. The process is presented as a sequence

of steps, from formulating the requirements to testing and deployment, where each

step is performed upon completion of the previous one (figure 6).

Figure 6: Implementation steps to develop a computer program for delivery to a customer (Royce, 1970)

System

requirements

Software

requirements

Analysis

Program

design

Coding

Testing

Operations

22

The problem in this model is that a possible change in requirements or an error on

earlier stage of the process undermines at least partially the progress that has been

made earlier, and requires returning several steps back in the process. Management of

changes is a conceptual problem of collaborative design that is directly related to

dependency of tasks. As suggested by the design theory discussed in section 2.2,

change is inherent in design. When the tasks are done in a sequence, changes to design

decisions made on early stages undermine related tasks that have been carried out later.

Design changes, therefore, imply the cost of re-designing.

Royce (1970) describes the impact that major changes in design have on software

development: “The required design changes are likely to be so disruptive that the

software requirements upon which the design is based and which provides the

rationale for everything are violated. Either the requirements must be modified, or a

substantial change in the design is required.” The second statement is similar to the

concept of reformulation suggested by Gero (1990).

On the project level, the cost of change for design project increases as the design

proceeds to completion, according to the study on construction projects cited by

Eastman et al. (2008), figure 7.

Figure 7: Cost of changes for design services (adopted from CURT, 2007, cited in Eastman et

al., 2008: 153)

Eppinger (2001) in his paper on product development projects mentions that there are

different types of dependency between project tasks. Some tasks can be put in

sequence, when a subsequent task utilises information from the previous task. Such

dependency produces feedforward information exchange. However, in creative

activities there are also tasks that are mutually dependant and require feedback loops,

when information from the later task causes change in a previous one, and causes

reiteration and rework. Such change loops are a distinguishing characteristic of

creative, innovative activities. (Eppinger, 2001)

Effort/

effect

Cost of change

Time

23

However, Eppinger (2001) states that many of the iterations that typically occur in

product development projects are marginally beneficial or wasteful, adding to project

costs due to rework they cause, and hence iterations must be managed. Typically it

involves shortening the iterations, since “keeping interdependent tasks separate can

cause considerable waste” (Eppinger, 2001). Royce’s example (1970), in which a

change on a later phase of a project undermines the whole project, is an extreme

example from Waterfall development where the length of one design iteration is equal

to the length of the project.

Software industry has eventually shifted largely from the Waterfall paradigm to Agile

methodology (Larman, 2004; Beck et al., 2001). In the foundation of this new

approach was the understanding of iterative nature of design and of the importance of

teamwork and collaboration. From the process perspective, most of the agile methods,

such as Scrum, aim at fixing design iterations to short time intervals and building

products incrementally, in steps during which the developers work in parallel on one

particular module or function of the product. On completion of each increment,

developers test the product and plan what should be done next, usually prioritising the

functions that have more importance or value for the customer. For example, the

Scrum method (Schwaber & Sutherland, 2013) involves short iterations called

‘sprints’, typically a few weeks long. Team meetings are held after each sprint, on

which the developers report on their progress and discuss the direction of their future

efforts.

The conceptual problems of collaborative design are also observed and described in

design-related literature from ‘heavy’ industrial domains, including manufacturing and

shipbuilding. The changes in design paradigm led by industry best practice and

enforced by scientific research on design have meant that the manufacturing-based

linear view on design process was largely discarded in the research. Similarly to a

study by Eppinger (2001) mentioned above, Parsons, Singer and Sauter (1999) in their

study on concurrent marine design highlight the iterative nature of design activity and

the coupling of design tasks:

“Planning. Design tasks cannot be sequenced in detail. There is generally no

way to progress through the design analysis and decision-making in an

organized way such that all the information is available when necessary to

each designer. Thus, the design spirals.

Coupling. Designers think locally, but they are tightly coupled with other

designers. Decisions that one designer makes affect the decisions that other

designers have made. This constant need for re-evaluation as a result of

changes in the design interfaces can lead to lengthy cycles of iteration and

change.

24

Prioritizing. Designers have no common language for comparing the

importance of issues.”

The concept of Concurrent Engineering (CE) was created in research on automotive

industry product development, and is a part of lean approach to product development.

Lean product development focuses on maximising the value and reducing the delivery

time through supplier involvement, integrated cross-functional teams, concurrent

(parallel) engineering and strategically-oriented management (Karlsson and Åhlström,

1996).

Concurrent Engineering (CE) is a type of design process where activities of different

disciplines are planned to be carried out in parallel. The purpose of CE is:

to reduce the delivery time (Bogus, Molenaar and Diekmann, 2005; Krishnan,

Eppinger and Whitney, 1995), and

to improve quality by increased collaboration and early involvement of

stakeholders who are specialised in production activities (e.g. subcontractors)

in design (Licker et al., 1996; de la Garza et al., 1994).

Overlapping the tasks can be a problem, if the dependencies between tasks are not

known. When the downstream activity has to start without the necessary information

from the upstream activity it has to be based on preliminary information and

assumptions that may change when the upstream activity is completed. The resulting

delay can eliminate the advantage of overlapping the tasks. This problem is called the

iterative overlapping problem. (Krishnan, Eppinger and Whitney, 1995; Eppinger,

2001; Bogus, Molenaar and Diekmann, 2005)

Therefore, the team must understand the effect of the design changes on downstream

activities (Krishnan, Eppinger, and Whitney, 1995; Bogus, Molenaar and Diekmann,

2005, Eppinger, 2001). The Design Structure Matrix (DSM) tool aims at reducing the

waste in product design projects and relies on identification of dependencies between

tasks based on assessment of the information needs of each task, in order to plan the

iterations and regroup tasks (Eppinger, 2001). When describing the application of

DSM, Eppinger (2001) mentions: “When we draw a DSM for a product development

process, we go to the grass roots and ask individual development teams what they need

from other teams to do their jobs.”

The Last Planner system of production control (Ballard, 2000) is based on a similar

approach. This planning and control system finds its application in AEC projects. The

system involves direct participation of project members in planning activities. First,

the project members work on understanding the interrelation of tasks and the

information needs of each task. Then they provide their estimates of how long it takes

25

them to finish the tasks. This results in a detailed schedule, which is reviewed

throughout the project.

Set-based design is an extension of concurrent engineering practice that addresses the

optimisation problem of design. The optimisation problem stems from the fact that the

design is limited by a number of constraints, such as budget and building space. This

problem relates to the interdependency of tasks. The constraints imposed on the

interrelated tasks mean that the customer value added by each of these tasks is

reversely proportional to the value of the other elements. Maximising value outcome

for one of them would decrease the value of others. An example of such relationship

is the limit of space in a car dashboard: air condition system, audio system, navigator

and controls conflict with each other in terms of space. The question is how the design

should be optimised so that it delivers the maximum overall value to the customer with

the given constraints.

In conventional point-based design the structure of critical and high-level elements is

determined and fixed at an early stage, and less critical or low-level systems are

selected to fit the high-level design. However, the elements that are given higher

priority in a project would not necessarily result in the highest value for the customer.

Prioritising individual tasks within a project leads to sub-optimisation of overall value.

In set-based CE (figure 8), first implemented in Toyota, the design process is organised

as a parallelised process of narrowing down the scope of alternatives for every system,

function or element in design, so that the optimal trade-off between them is achieved

(Licker et al., 1996; Parrish et al., 2007).

Figure 8: Parallel set narrowing process sketched by a Toyota manager (Parsons, Singer and

Sauter, 1999)

Time

Marketing concept

Styling

Product design

Components

Manufacturing system design

26

Application of concurrent engineering and set-based design to AEC industry has been

discussed broadly in the literature (Parrish et al., 2007; Anumba and Kamara, 2012; de

la Garza et al., 1994). However, for set-based design there was no empirical evidence

of application in the AEC in the literature, although there are no apparent barriers to

such. The general principles of CE, such as the emphasis on communication and

involvement of production site and sub-contractors in design, are relevant to the AEC

industry and can be recognised in the integrated approaches to AEC project delivery.

In addition to problems of collaborative design mentioned above, there are difficulties

for knowledge acquisition and dissemination in project-based collaborative activities.

Knowledge transfer and learning are generally difficult in project-based organisations

(Gann & Salter, 1998). In such firms, the design and production have temporary

organisation and result in one-off or highly customised products. Project process

usually causes discontinuities in the knowledge flows and learning. There is typically

a lack of connection between different business units of a project-based firm, and

project-based learning is isolated from company-wide business process. The lack of

organisational coherence in project-based firms hampers the accumulation of

knowledge. (Gann & Salter, 1998)

The problem with knowledge management is magnified when knowledge exchange

between different organisations is concerned. Traditional AEC practice, characterised

by ‘over-the-wall’ process, paper-based documentation, individual goals and restricted

information flow, does not support knowledge exchange and conversion (De la Garza

et al., 1994). Various design and production disciplines form knowledge silos.

Therefore, the emphasis on stakeholder integration and communication in the design-

related literature is also rationalised by facilitating knowledge transfer and learning.

The problems of knowledge conversion and exchange in AEC have a potential to slow

down the adoption of new technology, such as BIM.

To summarise, the methods of product development applied in different industries to

deal with conceptual problems of collaborative design include:

Project integration and early involvement of project stakeholders in

collaboration;

Meetings, face-to-face communication and shared workspaces;

Balanced iteration frequency and parallelism of tasks aimed at increasing

collaboration and mitigating the impact of design changes;

Involvement of teams in project planning and identifying dependencies

between tasks;

Set-based approach to optimisation of design value;

Systematic management of communication, requirements and knowledge.

27

The summary of directions for process improvement in collaborative design and the

list of corresponding literature sources are included in table 1.

Table 1: The possible directions for process improvement in design projects, based on the

reviewed literature

Direction

State

References

Undesirable Desirable

Concurrent process Sequential

workflow,

‘waterfall’ process

Concurrent

workflow; Managed

iterations;

Understanding task

dependencies

Royce, (1970), Parsons,

Singer and Sauter,

(1999), Parrish et al.,

(2007); Anumba and

Kamara, (2012); de la

Garza et al., (1994);

Bogus, Molenaar and

Diekmann, (2005);

Krishnan, Eppinger and

Whitney, (1995), Licker

et al., (1996), Eppinger

(2001)

Value optimisation Point-based design Set-based design Licker et al.,

(1996); de la Garza

et al., (1994)

Stakeholder

integration

No integration;

‘over-the-wall’

process

Early stakeholder

involvement;

Integrated delivery;

Trust and

collaboration

Licker et al., 1996;

de la Garza et al.,

1994, Aapaoja,

2014

Co-location Personal

workspace, few

meetings

Shared work space

or frequent meetings

Khanzode, 2013;

Eppinger (2001);

Kvan (2000)

Knowledge

management

Knowledge silos,

learning disrupts

after projects

Project-wide and

company-wide

knowledge sharing

De la Garza et al.,

(1994); Gann &

Salter, (1998)

2.5. Benefits of technology for collaborative design in construction

sector

The possible synergy between the information technology and the design process

improvement is not studied enough. However, the review of the literature regarding

BIM and other IT suggests that the information technology innovation has already

made and will continue to make a large impact on design.

28

The studies by Kleinsmann (2008; 2010) identify the variety of factors that affect the

development of shared understanding between project actors. The author categorises

them between three levels. The large share of the factors for achievement of shared

understanding relate to information, knowledge and communication.

For example, the second most frequent factor on the actor level identified in the study

was the ability to make transformation of knowledge when exchanging it between

disciplines. At the project level, the most frequent factor was the efficiency of

information processing. The examples of barriers related to it provided in the study

include late information delivery to the receiver, such that he cannot move on with the

next task, or that the receiver does not know the status of a document. The second most

occurring factor at this level was the quality of project documentation. (Kleinsmann,

2010)

Better information management in projects can reduce waste. The analysis of New

Civil Engineer (1991, cited by Baldwin, Thorpe and Carter, 1996) estimated that

around 25% of cost in construction projects carried out at that time could be eliminated

if construction information was managed effectively.

BIM, as an approach to management of building information in projects, is viewed as

a potentially disruptive innovation that can lead to major improvements. The impact

of BIM relates to the information process that underlies the physical construction

activity. The understanding of design as an information process and the overview of

conventional problems faced in design suggest that BIM can contribute to higher

efficiency and better quality of design. The benefits of BIM that can impact the design

stage of projects and are already utilised today include:

Visualisation. BIM gives even more advantages in relation to this general

benefit of computer-aided design: more efficient 3D design compared to CAD

and the integration of discipline-specific models can help evaluating design

options and solutions better (Eastman et al., 2008: 18). Visual channel is a

primary communication channel in design, because it can effectively represent

complex engineering objects and relationships that cannot be reduced to

unambiguous verbal descriptions (Maier, Eckert and Clarkson, 2005).

BIM as a design collaboration platform. The model shared by designers can

facilitate collaboration between different technical areas. It is possible to

exchange designs directly in the model, shifting away from traditional 2D-

based approach to data exchange. Using BIM as a multidisciplinary

collaboration platform can be possible with the help of Industry Foundation

Classes (IFC) and model server technology. (Plume and Mitchell, 2007;

Eastman et al., 2008: 19; Poerschke et al., 2010; Singh, Gu and Wang, 2011)

29

Clash and error detection. This benefit results from the possibility to

integrate discipline-specific design data. Savings from clash detection are

realised during construction phase of the projects. (Eastman et al., 2008: 19).

Cost estimation, quantity take-off and engineering analysis tools that use

data integrated in BIM, can support design development. Both cost estimations

and functional assessments can be faster and more precise with BIM, allowing

better evaluation of design alternatives starting from the earliest design phase

and enabling value optimisation in the project as a whole. Design analysis tools

can be used to improve energy efficiency and sustainability of buildings.

(Eastman et al., 2008: 18; Sattineni and Bradford, 2011)

As it was already mentioned, in spite of the possible benefits the technology still

suffers from software interoperability problems: different technical disciplines and

lifecycle phases require specific software solutions, and data transfer between software

is sometimes limited and can require substantial manual intervention. As Eastman et

al (2008: 15) explain, the companies can either use a set of interoperable software tools

from the same vendor, or procure software from different vendors that can exchange

data using industry standards. The problem of software interoperability is addressed

by such standards as Industry Foundation Classes (IFC), Land XML and GIS – the

common formats of data exchange developed by industry community or organisations

such as BuildingSMART for IFC, or National Geospatial-Intelligence Agency for GIS.

A relatively old study on electronic information exchange and alliance contracting by

Baldwin, Thorpe and Carter (1996) explores the issues of information management.

Although the set of technologies discussed in this paper is largely outdated (e.g. email),

the paper contains principles interesting in today’s context. The concept of electronic

exchange of 3D CAD data described in the article is relatively close to BIM. The

following benefits of electronic information exchange are identified (Baldwin, Thorpe

and Carter, 1996):

Guaranteed delivery of data

Improved quality of data

Reduced data handling

Improved communication between the alliance partners / stakeholders, and

Reduced risk of project delay.

The examples of interoperability problems, the delay of sending RFIs, the ‘over-the-

wall’ process and other inefficiencies discussed earlier show that, even 18 years after

Baldwin’s, Thrope’s and Carter’s study, there is still much room for improvement

concerning all the benefits suggested by them.

Despite the benefits of IT, physical communication between people is still essential.

The importance of face-to-face communication in collaborative design is highlighted

30

by the majority of reviewed studies and gives justification for such collaboration

methods as Scrum meetings (Schwaber & Sutherland, 2013) and co-location in Big

Room (Khanzode, 2012). According to Khanzode (2012), co-located teams are able to

design systems that better complement and support each other’s functions, and spent

less time on looking for up-to-date information or working with outdated information.

The potential role of technology is to aid co-working, for example, by providing visual

representations of the design during work meetings. Kiviniemi (2005) explores such

capabilities of visual technology in his study on decision-making during design phase

of construction projects.

Eppinger (2001) states that the exchange of information is a ‘lifeblood’ of design,

innovation activities. He adds: “When an electronics company's circuit designers know

what the casing designers are doing, they design a better fitting circuit for the casing.

And when the casing designers know what the circuit designers need, they design a

casing where it's easier to put in a better circuit.” This simple example, equally

relevant to other areas of collaborative design, can be used to outline the potential role

of BIM, and other IT, in maximising the value generated on design stage of AEC

projects.

2.6. Conclusions and introduction to empirical study

The review of the literature suggests that the benefits of BIM in the design phase of

construction projects have been studied considerably less than such in the construction

phase. The lack of literature directly addressing the topics related to design in the AEC

domain was one of the reasons why the literature review included studies from

different industries, such as software and manufacturing. This approach allowed

understanding different perspectives to the process improvement in collaborative

design. The similarity of principles that determine the efficiency and quality of design

according to studies from different domains shaped a theoretical background for this

research.

The key goal of this study is to identify the currently achieved and the potential benefits

of BIM for design projects in infrastructure sector. This search for benefits is to be

conducted via close examination of the current design process in several companies,

and by questioning the project teams about their perception of BIM and the design

process. This examination is expected to result not only in a review of benefits of BIM,

according to the research question 2, but also in the comparison of the current state of

design process to its potential state, and the recommendations of what improvement

steps can be taken by companies, regarding the technology or otherwise. These two

practical objectives are covered by the research questions 3 and 4.

RQ2. What is the potential role of BIM and other technologies in maximising the

efficiency and effectiveness of collaborative design in construction projects?

31

RQ3. What is the current state of design practice and BIM implementation in the

infrastructure sector in Finland? How do the professionals perceive it?

RQ4. What actions can be taken by the design companies in infrastructure sector

regarding the technological and process development in order to increase

performance in design activities?

The impact of BIM is to be studied on different aspects of designing. The identified

elements of process improvement that will guide this research include:

Concurrent process, project workflow and iterations, project planning

Optimisation of value delivered in the project

Stakeholder integration, contract-wise and information-wise

Co-location, meetings and other teamwork practices

Communication media

Knowledge management

Data exchange, information management

Some studies connect the implementation of BIM and other IT with the stakeholder

integration and relational methods of contracting (integrated project delivery). The

notion that stakeholder and cross-discipline integration is beneficial for the projects

was found in both the literature on AEC and on Concurrent Engineering in

manufacturing domain. Due to dominance of public contracts in infrastructure

development in Finland and the consequent legal barriers for using relational

contracting in infrastructure projects, it will be especially interesting to find out how

the possible directions of process improvement revealed in the literature study can be

applied to projects that use traditional procurement methods.

3. Methodology

3.1. Case studies

This research utilises case study methodology. The data for the empirical study was

collected in four design projects in infrastructure sector in Finland carried out by two

companies: two finished road projects (projects A and B) in Company Alpha, one

finished railway project (project X) in Company Beta and one unfinished railway

project (Project Y) in Company Beta. Projects B and X were conventional, and projects

A and Y used BIM: the contracts included creation of collaboration models and

delivery of models to the owner (table 2).

The finished projects were carried out several years prior to the study; therefore, a

substantial recall bias could occur if designers were asked about their perception of

past projects. Because of that, it was decided to obtain as much verifiable, documented

32

data for the past projects as possible (e.g. meeting notes, schedules, reports), and to

collect perceptions as to the date of the research without binding them to the case

projects. This way the focus of the study included not only the case projects but also

the case teams and companies on a longer term.

Table 2: Case projects

The division explains why a number of different methods were used in the research,

such as cross-functional project diagrams and questionnaire survey, and the

subsequent division in research data between case data and general data and

perceptions. A benefit from utilising different methods was that methodological

triangulation was achieved to enhance the credibility of results.

Prior to the study, semi-structured interviews were conducted with the managers of

each case project to obtain the initial information about the projects. The interviews

were audiotaped and transcribed.

The main part of the empirical study was to conduct workshops with project teams.

Two workshops were held for each project, at which project managers, BIM

coordinators and 2-3 key persons from different technical areas were able to discuss

the projects.

Because projects X and Y were performed by the same project management team,

most of the attendees were the same for the corresponding workshops. In addition,

these four workshops included an independent BIM expert from company Beta.

All workshops were conducted in a semi-structured manner, with some questions

being prepared in advance and some questions emerging in the discussion. The main

language was English, but Finnish was also used. Maila Suvanto from Vianova

Systems Finland Oy, who was an instructor of this thesis research, assisted with the

workshops and helped to translate the discussions. The workshops were videotaped,

the duration of each workshop was 3.5 hours.

A first and a second workshop were planned differently, and served different purpose:

the first one – to obtain general data, and the second – to conduct a detailed discussion.

Company Case

project

Case

team

Project

field

BIM Year

when

finished

Number of participants

1st

workshop

2nd

workshop

Alpha A 1 Road Yes 2013 4 4

B 2 Road 2010 4 4

Beta X

3 Railway 2013

7 5 Y Railway Yes Ongoing

33

On the first series of workshops the project teams provided general details about the

projects, described the project events and problems. They used post-it notes to create

maps of the projects that reflected the main project events, tasks and data exchange.

The secondary data was also collected:

Logs of received and created data; logs of design revisions

Project schedules, milestones

Meeting dates and agendas, meeting notes

Project diaries

Reports, “lessons learned” notes

To create the project maps, the teams were given coloured sticky notes and asked to

write down the various project activities and corresponding dates, using their laptops

to gather information right away. They were also asked to write their names or initials

on the notes. There were five colours in total, and each related to a particular type of

process step or event: ‘Task’, ‘Received item’, ‘Delivered item’, ‘Meeting’, and

‘Event’ (see examples on figure 9). The filled notes were then put on a canvas in a

chronological order.

Figure 9: Examples of note templates: ‘Delivered item’ (left) and ‘Meeting’ (right)

The project maps, workshop recordings and secondary data were analysed before the

second series of workshops. The data was used to create detailed swimlane (cross-

functional) process diagrams (Damielo, 2011: 6) of the projects, and to formulate the

agenda for the following meetings. The swimlane diagrams were presented to the

teams on the second workshops. The teams discussed the events and problems

discovered in the data and corrected a few inconsistencies in the diagrams. The

discussion was planned to test the hypotheses and cross-check the questionnaire data.

The broad discussion of different BIM-related and process-related topics allowed

drawing a picture of professionals’ perceptions of these topics.

The participants were asked to name the problems that are most important for their

work. Each team was asked to select 2-4 biggest problems, and every participant was

asked to take part in the selection and had his opinion heard. The named problems

34

were largely repeating the problems outlined in the questionnaire study that was held

in the beginning of first workshops. After selecting the top problems, the teams were

asked to split in pairs and discuss the reasons that were likely to cause these problems,

using the cause-and-effect ‘fishbone’ diagram, also called Ishikawa diagram by the

surname of its author (1950) (figure 10).

Figure 10: Fishbone (Ishikawa) diagram

In fact, many of the suggested ‘causes’ could be attributed to more than one category

at the same time, according to the designers (e.g. Method, and Process and Milieu, and

Environment). This was not a problem, since the purpose of the diagram was to serve

as an aid tool for discovering the problems and causal relationships rather than a tool

for categorising problems. At the end of the workshops, the professionals were asked

to write down the criteria of the hypothetical ‘ideal project’ as they see them, assuming

that unlimited resources were available.

3.2. Sentiment analysis

The method of sentiment analysis (Wilson, Wiebe & Hoffmann, 2008) was adopted in

the research to analyse the workshop recordings. The main purpose of using this

method was to support the credibility of results provided by qualitative methods, by

applying quantitative methodology to the textual data from the workshops.

The video recordings from the workshops were put on paper as narrations with time

markings. The narrations reflected each statement made during discussions by the

project team members, but were not verbatim. The researchers’ statements were

excluded. These narrations were analysed with spreadsheet tools. The results of this

analysis are presented in Chapter 4.

Problem

Man power/

people

Machine/

technology

Method/

process

Milieu/

environment

Material/

information

Measurement

Cause 1

Cause 2 Cause 3

35

For enabling the analysis of textual data, the data was transformed into spreadsheet

format: each recorded sentence or phrase with a complete meaning was transformed

into a short readable statement that reflected the original meaning non-redundantly,

but could be handled as a separate entity for quantitative analysis. The amount of

collected data allowed performing such transformation and analysis with spreadsheet

tools.

The analysis included two steps: the categorisation of statements and assigning the

sentiment index to the statements. The categories, or topics, were derived from

analysis of textual data and included:

Changes, requirements

Designing

People

Modelling/data

BIM implementation

Tools

Cost estimates

Clash detection

The categories were not exclusive. If a statement related to several categories – as

many categories as needed were assigned to this statement. Adding a new category for

the matter of analysis was easy – any category could be assigned to a statement if the

statement related to it, so the statement did not have to be excluded from the previous

category to be assigned to a new one. However, the categories were formulated in such

a way that most of the statements could be attributed to only single categories, and so

that the number of categories stayed reasonable. In this sense, the categories were

suggested by the data.

The approach was more flexible compared to if pre-defined categories derived from

theory were used, because the analysis was not restricted by older assumptions and

frameworks that had limited relation to the research data. Particularly, there was no

need for attempting to attribute any statements to categories that they did not strictly

belong to only because they went beyond the scope of theoretical frameworks.

In addition to being assigned with categories, each statement was marked with a

sentiment index: negative, positive or neutral (see table 3). A sentiment was given a

neutral index in most cases, and an index other than neutral was only assigned when

the attitude was explicit from the text (for example, if the word “problem” was

mentioned).

36

Table 3: Analysing the workshop recordings (fragment)

Item no. Original statement Short statement Sentiment

(1 – neg.,

2 – pos.)

67 09:08 Traffic signs budget somehow started to grow, and the PM noticed that, and it turned out that the traffic sign was taken to too detailed level, almost construction design level.

Traffic signs design was taken to too detailed level. The problem was noticed because the cost started to grow. (PM didn't see the designs?)

1

68 11-12 Construction costs were calculated both at the initial design stage and in road design, the number wasn’t much different, it was 50 M EUR.

2

69 13:20 Construction risks and construction cost estimate during design didn’t have to be considered that much on this stage, because all major decisions were made during general planning. So the estimates didn’t change that much.

All alternative options were considered during previous design stage, therefore construction cost estimate did not play a big role.

70 But generally, one big benefit that BIM can give is more accurate cost estimation while doing the design.

One big benefit that BIM can give is accurate cost estimation during design.

2

The limitation of such manual approach was that, compared to more complex methods

that employ mathematical algorithms, this approach to some degree was based on

interpretation. However, the work was carried out in such a way as to minimise the

impact of interpretation. As Chapter 4 shows, there is substantial amplitude of

variation between the results for different cases; in addition, the results of this analysis

are consistent with results suggested by other methods and lead to coherent

explanations.

Only semi-structured workshop discussions were included in the analysis. Discussions

of cause-and-effect diagrams and principles of ideal projects were not included

because they did not explicitly relate to the case projects, and because the double

inclusion of the same data in different analyses would undermine the results.

In the analysed discussion itself many statements did not relate to the cases alone;

however, excluding these statements would not be possible because the division

between case-related and general statements was not clear. Deciding upon such

division would add a large element of researcher’s assertion to the analysis. In

addition, there was a possibility of recall bias when designers discussed projects that

had happened several years ago.

37

Therefore, it must be noted that the workshop discussions and the results of sentiment

analysis are likely to represent the cases at least to some extent through the prism of

current state of work practice within the case companies at the time of the research.

However, the effect of recent experience on results is arguably low, because most of

the provided statements relate to cases and because the results of the sentiment analysis

are compatible with the unique characteristics of the case projects, which are also

reflected in other data.

3.3. Questionnaire

In the beginning of the first workshops, a paper questionnaire was given to the

attendees in order to obtain the individual perceptions and opinions. This allowed

collecting responses that were not affected by the authority of project managers or

project teams as a whole, or by a bias that could be introduced by researchers through

verbal discussion.

Open-ended questions were placed in the first part of the questionnaire to challenge

the initial research assumptions and minimise the influence of researchers. The

respondents were asked to state three problems that they deem most important in their

work practice, and list three most important benefits of BIM according to their

knowledge. This approach allowed discovering more issues than acknowledged by the

researchers when preparing the questionnaire.

These questions were followed by attitudinal questions, in which the respondents were

asked to rate their satisfaction in different aspects of work practice in their company.

The respondents were also asked to rate different aspects of BIM suggested by the

questionnaire. Bipolar rating scales of 7 points were used to ensure sufficient

distribution of answers. In addition, the respondents were asked to provide non-

attitudinal information on their work experience, education and BIM experience, and

to rate their knowledge of BIM.

The latter allowed evaluating the strength of the responses to BIM-related questions.

Two explicit ‘validation’ questions on knowledge of BIM were asked in different

places in the questionnaire: “How much do you know about BIM?” and “How well do

you know about the features of BIM, in general?” In addition, the respondents were

asked to provide the number of BIM projects in which they had participated. Cross-

checking between responses showed that this evaluation was valuable: the respondents

who reported no experience of working in BIM projects also reported less or no

confidence in their knowledge of BIM, and, most importantly, there were major

differences in responses to most of the questions between the groups with different

levels of BIM experience.

38

The low number of respondents (15 in total – all workshop attendees filled in the

questionnaire) would not normally allow generalising the questionnaire results outside

the case projects. However, the questionnaire data complemented and validated other

case data, and there were no contradictions between results provided by different

methods. The number of respondents is suitable for case study methodology chosen in

this research. Figure 11 represents the research data framework and specifies the

sources and types of collected data.

Figure 11: Data collected in the empirical study

4. Results of the empirical study

4.1. Questionnaire study

The questionnaire (appendix 1) distributed to the attendees in the beginning of each of

the first workshops included the questions about the attendees’ BIM experience, their

understanding of BIM functions and the perception of design process. All workshop

attendees have filled in the questionnaires. The respondents with no knowledge of BIM

whatsoever could skip the BIM-related questions. The designers were also asked about

their education and experience. Four designers had a Bachelor’s degree, and eleven

had a Master’s degree. The experience in the field was distributed as shown in table 4.

Table 4: Experience of workshops participants, reported in questionnaire.

Years of experience in the current field

Option

(years) b. 1-3 c. 3-5 d. 5-10 e. 10-20 f. >20

Respondents 2 2 5 1 5

There were differences in team member experience between projects. The participants

from Team 1 (project A) had 5-10 years of experience with the exception of one

participant with 1-3 years of experience, in that sense it was the ‘youngest’ set of

respondents. It was also a BIM project team. Team 2 (Project B) had 1 participant with

Interview Initial information about projects

Initial opinions (transcribed

audio)

Questionnaire

Individual opinions, perceptions

Individual information about

participants

Workshop 1

Project information

Project maps

Discussions (video,

notes)

Workshop 2 Project information

Swimlane diagrams

Fishbone diagrams

Principles of ideal

project

Discussions (video,

notes) Secondary data

Schedules, design

logs etc.

39

more than 10 years of experience and 3 participants with more than 20 years, which is

the highest experience among the cases. It was involved in a non-BIM project. Team

3 included two professionals with 3-5 and 5-10 years of experience, and one

respondent in each other category, being the most diverse project team in terms of

experience.

With regard to BIM experience, the designers could be divided in three groups, with

very straightforward borders between groups:

Group 1: five respondents with no experience in BIM

Group 2: four respondents who had 1-2 BIM projects – ‘novel users’

Group 3: six respondents who have reported participation in more than 7 BIM

projects – ‘BIM experts’.

The questions on BIM knowledge and experience were meant to evaluate the strength

and validity of responses to other BIM-related questions. The divisions between

groups with different BIM experience were significant. The participants’ level of

satisfaction with BIM (figure 12) and their willingness to work in BIM projects in

future (figure 13) depended on BIM experience and were highest for BIM experts.

Figure 12: The willingness to work in projects that use BIM, in relation to BIM experience

(note: the strength / credibility of answers may differ between groups).

Figure 13: The level of satisfaction with the process of working with BIM in relation to BIM

experience.

3,8

5,756,83

5,53

1

2

3

4

5

6

7

How much do you want to work in projects that use BIM in future?

No experience in BIM

1-2 BIM projects

>7 BIM projects

TOTAL

7: I want very much1: I don't want

2,75

4,5

1

2

3

4

5

6

7

If you have any experience in BIM, how satisfied were you with the process of working with BIM?

1-2 BIM projects

>7 BIM projects

40

BIM experts were more positive in their assessment of the effects of BIM on their

work. Figure 14 shows the responses to the matrix of scale questions, in which the

designers were asked to rate different aspects of BIM from very negative (-3) to very

positive (+3); the possibility to give “I don’t know” answer or no answer at all was

provided too.

Figure 14: The perception of the effect of BIM on different aspects of designing, in relation to

BIM experience (mean numbers).

The figure shows that novel users were more concerned about the difficulties of using

new technology. The impacts of BIM on (1.) time for design changes and additions,

(2.) project delivery time and (3.) ease and joy of working were rated the lowest by

novel BIM users, and BIM experts were neutral when stating their attitudes about the

effect of BIM on time (1. and 2.), acknowledging benefits with regard to ease and joy

of working (3.). Moreover, the analysis of questionnaire showed that Case A team has

viewed the impact of BIM on these three parameters most negatively compared to

other teams (for Team 1 mean numbers were -0.75, 0 and 0 for the three parameters

respectively, while the overall mean numbers for attendees were 0, 0.92 and 1).

The BIM experts, when describing the general reasoning of designers who resist to the

new technology, mentioned that many designers see BIM as ‘extra work’ (mentioned

-2,5 -1,5 -0,5 0,5 1,5 2,5

Awareness of project's state and actionsof other designers

Searching and finding the right data

Ease and joy of working

Project delivery time

Time required for design changes andadditions

Collaboration and information exchange

Visual exploration of design

Quality of design

Simulation, calculation and analysis

Clash and error detection

According to your understanding of BIM, rate the effect that BIM has on the following parameters (mean numbers):

No experience in BIM

1-2 BIM projects

>7 BIM projects

+3: Very positive-3: Very negative

41

on project A and Y workshops). The data presented in section 4.2.1 shows that the

concerns about attitudes and skills of individual designers were especially strong

among the respondents from Case A, in which designing and BIM were carried out as

separate processes by separate project management teams.

An interesting finding from Figure 14 is that the respondents with no experience in

BIM showed the similar general pattern of answers as those with experience, which

means that they were generally informed about the perceived impacts of BIM on

designing.

Figure 15 shows the standard deviation in the responses to the same question as

previous, but excluding respondents with no BIM experience. It shows that among

BIM users, the impact of BIM on ease and joy of working caused most disagreement,

whereas the effects on visualisation received similarly positive ratings from all

designers.

Figure 15: Standard deviations of the ratings of BIM’s effect on different aspects of designing

by BIM users.

More detailed analysis shows that within the group with experience of 1-2 BIM

projects, the impact of BIM on time required for design changes and additions was

rated as negative (= -1,25) unanimously, with the lowest standard deviation among all

groups (= 0,43). The standard deviations within groups with different levels of BIM

0,60

0,60

0,00

-0,47

-0,27

1,00

1,53

1,07

0,60

1,00

1,14

1,22

1,95

1,27

1,26

0,94

0,78

1,11

1,28

1,12

Awareness of project's state and actions of other designers

Searching and finding the right data

Ease and joy of working

Project delivery time

Time required for design changes and additions

Collaboration and information exchange

Visual exploration of design

Quality of design

Simulation, calculation and analysis

Clash and error detection

According to your understanding of BIM, rate the effect that BIM has on the following parameters (only BIM users included):

Standard deviation Mean number

42

experience were comparable, supporting the credibility of answers, with standard

deviations of the group with no experience being slightly higher for some parameters.

The most experienced BIM users were more aware of different features of BIM and

suggested more different features (Figure 16). The figure shows the results on analysis

of responses to an open-ended question asked in the early part of the questionnaire.

The question is formulated as following: “According to your understanding, what are

the most important features of BIM?” The attendees were asked to state 1 to 3 features.

When the responses were analysed and combined into the following figure, the

answers with identical meaning were counted under same categories, which are listed

in the figure.

Figure 16: Open-ended question about the features of BIM – answers categorised and

arranged in a stacked bar chart.

As the figure suggests, the most commonly recognised feature of BIM is clash

detection. BIM experts have suggested higher variety of features and recognised

process-related features more than the rest of the designers (e.g. ‘Process/work

efficiency’, 4th category from the top).

The figure also suggests that when the categories have similar meanings, BIM users

tend to mention more precise and specific attributes. For example, they prefer to state

0 1 2 3 4 5 6 7 8 9 10

Error/clash detection

Combining designs of different disciplines…

Data utilisation through lifecycle, data reuse

Process/work efficiency

Information exchange and transfer

Mass calculation

Easier change implementation

Machine control, prefabrication

Visualisation, 3D

Keeping data in one place, data administration

Reduction of unnecessary work, re-design

Automated creation of drawings

Collaboration and communication

Earlier utilisation of initial data

Increased amount of work

According to your understanding, what are the most important features of BIM? (open-ended question, accumulated chart)

No experience in BIM 1-2 BIM projects >7 BIM projects

Number of responses

43

‘error/clash detection’ over ‘combining the designs together’, and ‘data utilisation

through lifecycle’ rather than ‘information exchange and transfer’, although these

categories are interrelated (but not equal). Finally, the designers tend to name only

positive features, although one designer with no BIM experience have mentioned the

increased amount of work as the characteristic of BIM.

It must be noted that the questionnaire relates to the perceptions at the time of the

research and not to the case projects. Some of the responses might contain, in addition

to existing BIM attributes, some of the assumed or potential benefits that the designers

would like to see. However, in order to minimise this possible bias, it was stated

explicitly on the workshop and in the questionnaire itself that the designers only

needed to describe the current state of BIM in their companies.

In the next open-ended question (figure 17), the designers were asked to state the top

three problems that they face most frequently when working with other designers. The

question gave a view on the designer’s perception of process-related and collaborative

design problems. The results have been categorised in the same way as for the previous

figure.

Figure 17: Open-ended question about the problems of designing – answers categorised and

arranged in a stacked bar chart.

According to figure 17, the designers with more experience in BIM tended to suggest

issues related to IT skills, communication and information sharing significantly more

often than designers with little or no BIM experience. However, the latter suggested

0 1 2 3 4 5 6 7

Lack of communication

Lack of BIM and IT knowledge; resistance to change

Data transfer, format and interoperability problems

Keeping to the schedule

Lack of information sharing and collaboration

Working with outdated data

Unavailability of data

Lack of work coordination

Lack of process manuals

Managing changes

Conflicts in initial data

Differences in thinking and presumptions

What are the problems that you face most often when working with other designers and engineers? (open-ended question, accumulated chart)

No experience in BIM 1-2 BIM projects >7 BIM projects

Number of responses

44

more categories for unavailability of data and other data problems. The designers with

no BIM experience also stated much more the problems related to scheduling and work

coordination.

The figure does not necessarily suggest that in BIM projects some categories of

problems occur more frequently or cause more negative effect than in traditional

projects. A more probable interpretation is that designers with larger BIM experience

tend to deem other problems more important, compared to designers who have been

involved in none or few BIM projects. These might be equally the problems that BIM

is supposed to cope with, or the aspects of work on which BIM imposes higher

requirements.

The presence of some technology-related and BIM-related concepts in a question that

does not particularly relate to them was partly due to the workshop agenda, which was

communicated to the designers before the workshops, and partly to the fact that a few

BIM-related questions were placed earlier in the questionnaire. This influence on

answers was impossible to avoid completely.

The open questions ask the designers to mention 1-3 most important issues, which

means that the designers, even if they knew about more issues, would state only the

most important ones from their point of view. The figures 16 and 17, therefore, do not

attempt to show the difference in knowledge between designers (though that may also

be reflected in the answers), but primarily the differences in formulation and in

opinions on what is more important.

The categories of answers listed in figure 16 and figure 17 largely repeat the results of

workshop discussions and the problems of designing suggested at the end of

workshops, described in sections 4.2 - 4.4. They are also entirely consistent with the

problems outlined in the design-related literature that was reviewed in chapter 2,

repeating all categories described in the literature.

4.2. Case project A

4.2.1. Description

Case project A was a BIM piloting project undertaken on the general design stage of

a road and bridge project. It included reconstruction of a road and construction of

bridges. The case design project was special in two ways. First of all, a separate

contract was made for delivering BIM. The models were required to be delivered

alongside the standard paper-based documentation. The project team that attended the

workshop was responsible for BIM project only and included a BIM project manager,

a BIM coordinator and two modelling specialists from road and bridge disciplines. The

responsibilities of the team during the project included developing the BIM

45

requirements together with the owner, coordinating the modelling work and creating

the BIM deliverables. The case team had limited involvement in actual designing.

The second reason why the project could be considered special was that the initial

design stage that preceded the project had resulted in relatively detailed design

descriptions, which had taken the initial design almost to the level of detail of general

design. The majority of design solutions had already been defined before the project,

as well as the approximate corridor in which the road was going to be built; this was

related to the fact that part of the road went through the town area with private

buildings and land, and negotiations with landowners had to take place during the

initial design stage.

The main design project included up to 20 designers. The cost of the design project as

incurred by the owner was 500 000 EUR, the estimated cost of the whole project,

including design and construction – 50 million EUR. The duration of Case project A

was 10 months. The project, both design and modelling contracts, was procured with

DBB method.

BIM in the project was implemented on both the preceding initial stage and the general

design stage. The first goal for the owner was to test the capabilities of BIM and use

the models in construction phase. The created initial and collaboration models were

used for clash detection during design. In the conceptual design phase, models were

also used for comparing the design alternatives. Finally, the piloting project was used

by the owner to develop BIM requirements and modelling standards that could be used

in future projects, and to acquire higher expertise in BIM.

As the case project began, the team held a series of meetings with the customer to

define the scope of their work and BIM requirements for general design. They started

with the initial meeting of the steering group, which included BIM experts, managers,

representatives of different technical fields and customer representatives.

After that, the company organised meetings of BIM groups within each of the major

disciplines: road design, bridge design, geotechnical design, and road signs and

lighting design. The purpose of these meetings was to formulate the modelling

requirements discipline-wise. The series of meetings ended with the second steering

group meeting that confirmed the requirements and approved the budget estimated for

BIM project.

As already mentioned, the customer still required all traditional 2D documentation to

be delivered. When discussing the benefits of BIM on the workshop, the project

manager and the model coordinator have told that design cannot be made entirely in

3D in any foreseeable future. The reason for this is that 2D representations in many

cases have better visualisation properties and are easier to understand and navigate;

46

they can carry technical data that would be hard to display in a 3D model, and therefore

are more usable for detailed design. This was mentioned with regard to both the idea

of designing entirely in 3D and replacing 2D documents with 3D models. However,

some 2D forms required by today’s standards could be replaced by 3D, according to

the attendees.

4.2.2. Workshop discussion

The issues discussed on the workshops are presented below. The team members have

described the case project and expressed their views on different problems related to

BIM implementation. The discussion on the Case A workshops was largely focused

on the process aspects of BIM, the benefits of BIM in the case project and in general,

and the problems encountered in the case project regarding process, people and

technology. The team generally did not discuss designing and issues related to

designing because the team was not in charge of the actual design work.

Modelling and data

The BIM project started with preparing the initial model using the data provided by

the third party. The initial model included old highway, surrounding structures, terrain

model and other major elements of road environment as they were before the

construction.

The initial model was used for detecting clashes in initial data and for serving as a

basis for collaboration model (the collaboration model included the initial model, so

that new designs could be compared to the existing state of the site). The team reported

major problems with the quality of initial data, despite the fact that the third party had

created their own initial model that should have been used to reduce the amount of

errors and inconsistencies in the data. The revealed problems were fixed mainly by the

model coordinator and additional data had to be requested too. Company Alpha also

had its own old model created during the previous design stage, and eventually it was

used too. The resulting initial model was not used by most of the disciplines in the

main design project, e.g. the bridge designers only used the standard 2D

documentation.

Construction of collaboration model and preparation of BIM deliverables was carried

out mostly at the later phase of design project, when the majority of designs were

finished. The data from different disciplines was added to collaboration model by the

model coordinator each time the corresponding parts of design were ready. Every time,

an examination for errors was carried out and designers were instructed what new

changes and fixes should be made to the data.

The model coordinator implemented small changes to models to make the designs fit

each other in the collaboration model, so that to avoid the lengthy process of sending

47

instructions to designers and waiting for new data. As the model coordinator told, most

of the data delivered by designers had to be modified at least to some extent before it

could be integrated. In case of geotechnical design, which was outsourced to a third

party, the required changes were substantial. There were two errors in provided

geotechnical data that required the designs to be sent back for correction, and many

changes still had to be made by the coordinator afterwards.

When describing the modelling issues in case A, the model coordinator stated: “(There

were) problems with geotechnical models, mostly with the quality of the models, so I

needed to modify them… Retain walls I made by myself from 3D designs…

Foundations of the bridge were all over the place above the ground.” The team have

experienced problems related to data formats and software interoperability, and

problems with coordinate systems.

The exceptions from these problems were road design that provided 3D data of good

quality, and bridge design, in which a designer qualified in modelling was involved to

create 3D models from standard 2D documentation, since the bridge design itself was

done in 2D. The manual transition of bridge design to 3D was a difficult and time-

consuming task, and the design tool used for the task had limitations with regard to

modelling complex geometries. But the created bridge models did not have to be

modified by the model coordinator.

The process of revision and correction of collaboration model continued for as long as

three months. The collaboration model was also reviewed on meetings. According to

the team, the designers did not use collaboration model in their work, so the model

was only useful for the client.

Another example of problems faced by the team related to miscommunication of

instructions to one of the designers. The specialist took the modelling to a level of

detail much higher than that required by the owner. This was unnecessary work. The

problem was noticed by the project manager only when the costs related to this

discipline (costs were calculated based on spent man-hours) became noticeably higher

than they should have been.

Cost estimation and clash detection

Because the main structure of design was defined and decisions on alternatives made

in the initial design phase, cost estimation did not play an important role in the case

project. It was done in a conventional way, using cost calculation software with manual

data entry, and was carried out at the later stage of the project. However, the team

believed that automated and more exact cost estimation while doing the design was

one of the major potential benefits of BIM.

48

Clash detection using the models was done by the model coordinator on each revision

of the initial model and collaboration model, when new data was added. It was done

manually by observing the designs, although some solution for automated clash

detection was tested at one time during the project. After revisions the information on

the detected clashes was sent for the designers for correction.

Model-based clash detection was one of the main benefits of BIM for the project. It

revealed two major clashes that otherwise would not likely be found. One of the

clashes, for example, was discovered with the help of initial model and related to

pipelines that were planned to be built in the area of construction. The third party that

had designed the pipelines was contacted and the conflict was resolved by changing

the pipes’ location. As the customer reported to the design team, the potential cost of

re-work related to the detected clashes exceeded the cost of BIM piloting, making the

piloting project fully pay off.

Planning

The Team 1 emphasised the role of the main project manager in planning and

coordinating the design, especially at the starting phase. The project was mostly

planned by a project manager according to his experience, and the experts from

different technical areas were taken one by one to start the work. As the team

explained, the design projects cannot involve too many people and spend too many

man-hours on planning, as it would increase costs. Therefore, project manager had the

leading role in the beginning of the project. The team wished there would be enough

time in projects to involve more people in the initial phase. However, in most of the

projects it was not possible, according to the team.

In spite of multiple difficulties in modelling, the team was able to meet the budget.

The design project was delayed, generally due to the schedule of the customer, and the

change of deadline was approved by the customer. According to the model

coordinator, this was positive for the BIM project, because it gave more time for

modelling.

People and tools

The team has reported on the workshop that communicating the modelling instructions

to designers and making designers deliver the right data was a difficult task. The

project manager and model coordinator had noticed designers’ resistance to change

during the project, and were generally dissatisfied with the IT skills of older designers.

During the workshops, the team has mentioned some deficiencies of currently

available tools as one of the reasons why motivation of designers to adapt was not

high. Some deficiencies of tools could also explain the fact that designers almost did

not use the initial and collaboration models. For example, according to the team, the

49

navigation problems that existed in the software made it hard for common designers

to learn how to navigate the collaboration model in order to use it for clash detection.

Therefore, clash detection between different disciplines relied largely on the model

coordinator. In spite of the improvements, according to the members, “…the software

is not yet on the BIM level”, although it was admitted on the workshop that some

newest tools might have better functionality than that in the case project.

The team emphasised that the software used at the time of case project did not take

into account some of the requirements of designers. As one of the members mentioned,

“The developers do not know designing well enough on the grass root level”.

According to the team, it was important for software industry to develop BIM tools

that would be more usable and would give visible benefits to designers in their

individual work.

Sentiment analysis

Figure 18 shows the results of the sentiment analysis for Case project A. As the figure

shows, the most commonly discussed issue by the project team is modelling/data,

particularly, low quality of initial data and problems with designs delivered to model

coordinator during the project. Second and third most problematic categories are

designers’ attitudes and IT skills (people), and tools. The dominance of topics related

to BIM in the discussion can be explained by the character of the case project and the

case team.

Figure 18: The occurrence of topics and sentiments in workshop discussion: Case A.

0 2 4 6 8 10 12 14 16 18 20

Collab. Model

Clash/error detection

Cost estimates

Tools

BIM implementation

Modelling /data

People

Collaboration

Project planning

Contracting

Design

Changes, requirements

Sentiment analysis, Team 1

Negative sentiment

Neutral sentiment

Positive sentiment

Number of statements

50

The domination of negative sentiments in the topics points at what kinds of problems

this team have experienced in the case project and possibly in other BIM projects. The

overall picture is especially noticeable when comparing Figure 16 to the Figure 18 that

represents the Team 3, which was involved in BIM project Y and included experts

with comparable level of BIM experience, but was more positive in the statements.

These results do not suggest that BIM did not have value for the Case A or that BIM

in general was perceived negatively - this would contradict the team’s statements and

the questionnaire results. Rather, the figure suggests that the team has acknowledged

the difficulties that affected their work in project A, particularly on the process level.

Summary

The results of the Case A analysis outline that the team of BIM experts in the case

project faced numerous problems. The discussion with Team 1 revealed substantially

more negative sentiments than the discussions with Teams 2 and 3. The following

issues are perceived most negatively by the team members:

Low quality of initial data required correction and completion efforts

Resistance of designers to adopt new methods and comply to guidelines

Designers’ lack of software skills

Shortcomings in data provided by designers and third parties required

comprehensive correction

Capabilities of tools were ‘not on BIM level’

The problems related to working with designers can be attributed to the knowledge

gap between BIM experts and new BIM users (most of the designers are either new

users or do not use BIM). The questionnaire has already shown that the new users of

BIM are prone to view BIM more negatively and see it as more time-consuming, since

they do not have enough skills.

In project A, one of the facilitators for the division between BIM experts and common

designers was the fact that modelling was largely disconnected from the design

process. This division was even fixed in the contract. Model integration and correction

mostly relied on model coordinator, and the BIM-team had to deal with instructing

designers on modelling and corrections. The team agreed that the designers in the

project “followed instructions at most”.

The diagram of the project showed that the project correction and completion phase,

which was carried out after most of the design work was finished, took as long as three

months to complete. This might suggest that modelling was in some ways a bottleneck

for project delivery, although this is hard to verify. The benefits of BIM for

collaboration were not realised as the designers did not use the collaboration model,

and the model was mostly created at the late phase of the project. Having modelling

51

and designing in the same project, having one project management team for the whole

project and “designing by modelling, from day 1” was mentioned on the workshop by

the team members as something they would like to look for in future projects.

Despite the problems, the piloting project was beneficial for the owner. First of all, the

owner reported that two clashes identified during the project using the initial and

collaboration models could cover the cost of the whole BIM piloting, since they

constituted major risks for construction. Secondly, the project involved developing

BIM requirements and guidelines, and has contributed to BIM expertise of both the

client and the design company. As the team reported, the expertise of Company Alpha

in BIM initially exceeded that of the client, and BIM experts from the company had a

major role in formulating the BIM requirements.

In general, the team was positive about BIM as a developing technology that has a

potential to facilitate the work of design companies in infrastructure sector, although

many prospect benefits of BIM were not yet realised. The team emphasised the role of

process, people and management in achieving improvements in the sector, and their

primary role in successful technology change.

4.2.3. Problems of design projects, Team 1

At the end of the workshop, the team was asked to select the most important problems

that the designers face in their work. The problems were largely suggested by the

previous discussion, but do not refer to only the case project. After the two key

problems were selected, the team was asked to name the causes for these problems

using the Ishikawa diagram as an aid tool. The two problems and their causes stated

by Team 1 are:

“Lack of collaboration

o Design and modelling is separate

o Lack of tools for data management

o Old habits of designers

o Lack of IT skills

o Lack of information sharing

o Collaboration model is under-used

Faulty initial data

o Lack of interoperability

o Lack of common formats

o Lack of quality checking

o Lack of process guidelines”

The named issues relate to the same categories as those that caused the most negative

sentiments during the earlier discussion: resistance to change among individual

52

designers and problems with modelling and data, such as lack of interoperability.

Particularly important is that the discussion of technology goes beyond BIM: the team

mentioned the lack of data management tools as one of the problems. In Case project

A, project bank was used as a tool to share finished designs; however, most of design

information was exchanged through internal project drive and even email, which could

sometimes lead to delays in data delivery, dispersal of information between many

sources and the inability to find the right data when required. It was not clear from the

discussion how high the degree of impact of this problem was on the team’s work.

In general, the central topics of the Case A discussions were modelling, information

process and communication. The relation between process improvement and

technology improvement was constantly mentioned by the team members. The

members tended to view process as the primary issue, and technology as the facilitator

of design, collaboration and management. As the project manager mentioned, “BIM

has much potential to make things more effective, but the key is in the process and the

people running this process, and the key is always communication, and knowing in

advance what is supposed to happen. BIM is a tool; it can be a tool to support project

management.”

When describing the hypothetical ‘ideal’ project, the members stated the following

principles:

“Right and appropriate tools for design, project management and

collaboration

o Everybody knows how to use them

o Software are easy to use

Clear guidelines both in design company and client organisation, and

commitment to guidelines

Communication

Right attitude

Right people

Goal-oriented customer

Customer who has knowledge of BIM

One project manager and one project (design and BIM are not separate)

Modelling starts from day 1, design is done by modelling

Collaboration meetings with all disciplines every two weeks or when needed

Accurate schedule

Initial data received in the right format

Efficiency: everybody knows what they are doing (right things in right order)

Critical self-evaluation after every milestone/project/design stage

Continuous development of knowledge and process; systematic learning”

53

The team’s vision of an ideal project highlights some of the same issues as the

designers have expressed during the workshop, and, as the following sections will

show, largely repeat the issues that other teams participating in the research have told.

In addition, the project manager expressed his opinion about the large size of BIM

piloting projects. According to him, implementing BIM in a large project makes it

difficult to achieve many benefits and is not as effective as investing the same money

in a smaller project could be. Therefore, according to his opinion, public owners need

to conduct smaller piloting projects where it would be possible to reengineer the entire

process and take it to the actual BIM level.

4.3. Case project B

4.3.1. Description

Case project B was a conventional design project for road reconstruction. The project

included the increase in capacity of an existing road and the construction of new

interchanges. Case B was a road planning project. Road planning is commonly a first

stage of road design process in Finland, and when it is, it incorporates both the initial

and general design. It results in the development of a road plan – a set of documents

including maps and cross-sections that has a medium level of detail and is used as

input data for construction-level design.

Like in all case projects, DBB procurement method was used in Case B. The ownership

was divided between two public authorities, and a number of other stakeholders (e.g.

other facilities in the area of the site) participated in the project imposing extra

requirements and providing the initial data. The cost estimate for the construction work

was 42 million EUR, which is comparable to Case project A. The cost of the design

project was 703 000 EUR and the duration was 18 months, which is more than in Case

A, and is related to the fact that project B comprised both the initial and general design

stages.

4.3.2. Workshop discussion

The content of the discussion with Team 2 significantly differed from that with the

team 1. Team 2 discussed primarily the evolvement of design and the common practice

in traditional design projects. The members explained in detail how the design is

developed by technical areas and how the requirements and cost restrictions are

embodied in design. The effect of changes in requirements on the design process was

also discussed.

Another group of topics discussed during the workshops was related to planning,

coordination of resources and coordination of work between technical areas. The

designers described problems that stem from the constant lack of resources and

54

unpredictability inherent in project-based activities, as well as from the common

shortcomings in the industry practice. The discussion also included the issues related

to communication and data exchange, and outlined the existing approach to

collaboration in the traditional projects.

Design and data

As the team described, the case project started with the initial data gathering and

meetings with owner representatives. The initial data included terrain data, older

geotechnical data, drawings of existing road, structures and engineering systems, land

use plans etc. Requesting for additional data was mostly the responsibility of

individual designers. Additional geotechnical surveys were carried out by the company

and the results were significantly different from the existing data.

The design process in road projects is generally point-based. The design is defined by

existing environment and the planned road capacity, which is the critical requirement

for design. This requirement is communicated in middle-term and long-term traffic

forecasts. The forecasts are used by designers to make traffic simulations, which

determine the road geometry, number of lanes and types of ramps and intersections.

Traffic simulations in the Case B were done using the simulation software and the

proposed traffic solutions were assessed by road designers. The alternatives, when

there were such, were shown to the client. This process of conceptual design reiterated

until basic plan of the road was created.

The team members told that there were no problems related to the quality of initial

data. According to them, the more usual problem related to initial data in their projects

is incompleteness: some data can be missing and has to be requested again.

Project bank was used to deliver the finished designs. The data was exchanged

internally using the project drive. The designers were instructed to always save the

updates to the drive, instead of storing them on computers. As the project manager

have said, the designers should not (and in most cases do not) keep the newest files on

their own computers, because it can disrupt the work of other designers by making

them use old or incorrect data.

Keeping all files up-to-date was considered necessary by the team; it ensured that all

designers and the project manager had the correct and latest information. However, as

the members have told, in some cases a designer is unaware that a change has occurred

in another person’s work that affected his own work. For example, if the change is

mentioned on a meeting and the designer affected by this change is absent, the issue

might be forgotten inside the meeting notes and not communicated to the designer,

causing him to unknowingly use the incorrect data or assumptions, and eventually do

extra work to correct the designs after the change is noticed.

55

According to the team, working with outdated information is nevertheless a rare

problem because both the creator and the user of data usually check that it is up-to-

date (the user sometimes have to ask if a shared file is newest). As the members said,

it is always important that the designers discuss their work face-to-face, not only by

sending the drawings, because there are some issues that cannot be overlooked without

direct communication.

Requirements and changes

One of the problems in the case project was that the scope of design was not defined

before the project, thus the pre-design was done partly after designing had started.

Changes in the scope during design required additional orders from the owner.

Normally, if additional work is required due to change of customer requirements or

change in proposed solutions (e.g. construction of additional bridge), the deadline for

the project can be moved under an agreement with the client. In the case project there

were additional orders but the schedule was not changed. As the designers believed,

the fact that pre-design overlapped with road design was beneficial for the quality of

developed solutions.

According to the team, the general impact of design changes is the highest for

disciplines that start later, such as bridge design, because they often have to work with

partially ready data from road and geotechnical design in order to meet the schedules.

This is consistent with the literature reviewed in section 2.4. The changes in the data

are more likely to occur when designers work with data from tasks that have not been

completed. These changes in data can make some of their designs become obsolete

and result in re-design. As the bridge designer explained, many years ago it was

possible for bridge designers to start when the preceding design tasks were completed,

but nowadays it was not possible because the schedules became narrower.

Cost estimation and clash detection (conventional)

The cost of design options was assessed whenever any alternatives were considered.

In the case project most of the alternatives related to geotechnical design. Cost

assessment was done using specialised software, with manual data entry.

The costs in projects are usually calculated only for the fragments of design that qre

affected by the alternatives. For example, when a new bridge option requires different

road geometry, all impacts of this option on road construction costs are also included

in calculation. According to the designers, because the costs are calculated by the

specialists from the corresponding technical areas, it is important to agree who

calculates what so that same costs are not calculated twice.

Clash detection in the project was carried out in a traditional way. This means that

designers from different technical areas had to exchange and cross-check drawings and

conduct face-to-face meetings in order to ensure that designs fit each other. Designers

56

usually review their drawings in pairs by comparing them against each other, placing

one drawing on top of another in design software. At the end of projects overall clash

detection is usually done including all disciplines.

In the case project face-to-face meetings were easy because all designers worked in

the same office. According to the team, face-to-face meetings for ensuring design

concurrence are easier for those designers who work on the same floor than for those

on different floors. Regular project meetings with the whole team also help clash

detection: the meetings make designers aware of what is happening in other disciplines

and provide them with information on where to look for clashes.

Besides finding the clashes between disciplines, the designers have to ensure that their

own designs are correct and consistent. The designers create many cross-sections,

including alternative solutions that are not necessarily included in final design. When

working with 2D, the designers have to move between different projections, from XY

to XZ to YZ, and make all views consistent with each other.

Project planning and collaboration

The designers had substantial impact on planning the work, although the project

manager had a leading role in planning. The designers made their input in scheduling

by giving estimates of when they could start and how much time they needed to

complete their tasks.

As the team members explained, planning in the beginning of the projects is usually

done forwards, so that the tasks are arranged starting from the beginning; later, when

the content of work is known and the deadlines are set, the tasks are rather planned

backwards starting from deadlines. This was reflected in the case project’s time

schedule, in which all tasks started at different points of time but finished at the same

time. A common problem for late design disciplines, such as bridge design, is that

sometimes they do not have enough time to finish their work before deadline because

previous tasks has taken too long and has delayed the start of their work.

In addition, later disciplines such as bridge design have to take part in early stages of

design when their participation affects the earlier disciplines. Their involvement in

early design phase is needed in specific cases, where multiple options are available

and where some problem or risk exists regarding their technical area. For example, the

requirements of bridge design can affect the road design: if one road goes above the

other road in a too narrow angle, specific road geometry or other solution may be

required to ensure that the bridge can be constructed. This way the requirements of

later disciplines are realised in earlier phases of design, to ensure constructability.

57

Generally, the team members mentioned that virtually all disciplines are connected

and can have some impact on each other. The team emphasised the importance of

communication and regular meetings in projects.

In the project meetings were held monthly, in order to review the designs and discuss

the progress with the owners. On these meetings the team discussed what had been

done and what should be done before next meeting. The team discussed changes and

provided information for the customer to make decisions. On each meeting, the agenda

and date of a next meeting were set. In addition, the company held special meetings

with other stakeholders such as public organisations, land users and third parties (e.g.

meetings with electricity grid operator). A public meeting was held to present the

design to stakeholders and public.

According to the team members, it is usually difficult to plan the projects and

coordinate the work. Designers work with many projects at the same time. The

schedules are usually not precise and are done according to experience. The intensity

of work within a project can change in relation to other ongoing projects: designers

generally start spending more time weekly on one project compared to other projects

when the deadlines are approaching. The members have stated that the work for each

project becomes more intensive before each project meeting, and the intensity

generally increases towards the main deadline.

There is no general schedule that could combine schedules of all projects of the

company. Different projects are coordinated on high level by project managers, but

this coordination is limited, even though they might share the same designers. Usually

the company competes for all tenders, even though the designers might be occupied

with existing projects. If two projects have to be done at the same time, the more

important project is prioritised and the less important one has to be delayed. The

tenders and projects are seasonal, because designs usually have to be ready in spring

before the start of construction season.

The designers have repeatedly mentioned that their time is always scarce, and it is

usually very hard to keep to the schedules. Because of that, they cannot invest too

much time in planning in the beginning of projects. In fact, designers are always

occupied with their work. According to the project manager, planning and project

management is much more basic and simplistic in infrastructure engineering than in

other industries, and “schedule is only an estimate” in infrastructure design projects.

Sentiment analysis

The analysis of textual data from the Case B workshops revealed substantial

differences from the Case A (Figure 19). The number of negative statements in the

discussion was considerably lower than in the previous case. People and tools – the

58

topics that had caused overwhelmingly negative sentiments in Case A – were almost

not discussed in Case B workshops. The amount of neutral statements was high in

project B.

Figure 19: The occurrence of topics and sentiments in workshop discussion: Case B.

The discussion was not focused on modelling or technology, although computer-aided

designing and design data were among the discussed issues. The most frequent topics

were collaboration, project planning and design. Topics exclusively related to BIM

were not discussed. Even though BIM was mentioned in the workshop agenda, the

members of Team 2 had limited knowledge about BIM and the discussion was focused

on other issues.

Summary

The study of the case project provided a comprehensive view on design practice in the

infrastructure sector. Process-related and collaboration-related issues dominated the

discussion. The discussion had the lowest number of critical sentiments, and many

issues that caused the criticism in case A were not discussed by the case B team.

The traditional design process included a number of good practices, such as systematic

information sharing and involvement of designers in planning. However, the team

outlined a number of problems in design activity, which largely mirrored the

conceptual problems of collaborative design identified earlier in literature review.

The issues of project planning and resources draw particular attention. The limitation

of resources and common industry practice was viewed by the team as the main reason

0 2 4 6 8 10 12 14 16 18 20

Collab. Model

Clash/error detection

Cost estimates

Tools

BIM implementation

Modelling /data

People

Collaboration

Project planning

Contracting

Design

Changes, requirements

Sentiment Analysis, Team 2

Negative sentiment

Neutral sentiment

Positive sentiment

Number of statements

59

for inaccurate and basic project planning, both in terms of scheduling accuracy and

definition of design content in the beginning of the project.

Another planning-related problem outlined by the team, which has also been

mentioned by Team 3 in the following case, is that designers participate in many

projects at the same time. There is no general schedule that could integrate all project

schedules in the company. The Team 2 members suggested that right project

management tools and software could possibly help planning and coordinating the

project, but such tools were not available. The members repeatedly mentioned face-

to-face communication, work coordination and systematic data handling as important

facilitators of efficient design process.

4.3.3. General problems and the ideal project – Team 2

The first problem selected for discussion was management of change in projects. The

changes in schedules were in focus of discussion, but changes in work content and

design were discussed too, as they typically affect the work progress and might cause

delays.

According to the team, possible changes are often not considered by project

management and their impact is not taken seriously. When designers start the project

they know approximately when they have time for it with regard to other on-going

projects. A change in the schedule that moves the start of the tasks to the points of time

when designers are busy with other projects means that at least one of the projects has

to be delayed.

Every change can cause a chain reaction, for example, a delay earlier in the project can

affect the bridge design, and the designers will have much less time to complete their

work. If the possibility of delay is not thought in advance and the bridge designers do

not have available resources when needed, the project will likely miss deadlines or the

bridge designers will have to compromise on other projects.

According to the members, changes always occur and are not always possible to

oversee, but a project team and a project manager should never stop thinking of risks

and always consider what is going to happen next, in order to take preventive

measures. When stating the issues that can cause problems with change management

the team listed both communication problems and technology. Some points related to

the problem of checking other disciplines’ work and achieving awareness of the

disciplines’ progress:

“Change management problems

o Lack of software skills

o Too tight schedules

60

o Lack of communication

o Lack of cross-checking between disciplines

o Designers use different formats/templates

o Customer has too tight decision schedule”

The last issue on the list relates to restrictions that the customer’s decision-making

schedule imposes on projects. Usually no changes in decision-making schedule are

accepted by a customer, so in case of delay in a project, the quality of design would be

compromised rather than the customer’s decision schedule. For example, in one of the

projects the client was ready to give up some of the previously required design

documentation in order to preserve the dates of decision making.

The second problem related to project planning. According to Team 2, project planning

is done on too high level, meaning that planning is very general and there is not enough

accuracy:

“Project planning is done on too high level / not accurate

o Amount of resources hard to estimate

o Lack of tools (e.g. software) for project management

o Old practices of planning

o The inputs from disciplines are inaccurate

o Not enough information from contracting phase”

As the team members said, schedule is mostly an estimate. Typically, when different

design disciplines provide their design plans and time estimates to project managers

for making the schedule, project managers cut the suggested timeframe by 30%

because the timeframe is too long and makes the project too costly. After that,

according to Team 2, “nobody plans how to design it with only 70% of time”.

The members have also mentioned the lack of tools for project management. This

problem might have several reasons: the company not using the tools, the project

managers being unfamiliar with available tools, or existing project management

software being inappropriate for the needs of infrastructure design projects. The

reasons were not broadly discussed with the team.

The last problem suggested by the team members was not having enough resources in

the beginning of design projects to ‘design the design’ – to define the design content

and formulate the requirements for design process and output.

“Not enough resources for designing the design

o Industry practices do not support proper project management

o The design detail/accuracy should be decided in advance

61

o Project organisations grow large but do not have general/master

planning

o Competitive bidding prices do not allow designing the design

o Customer’s schedule has unrealistic expectations

o One person does not have necessary skills and knowledge to design the

design

o There is a tendency in consulting firms to pass the extra work to other

consultants when the schedules are too tight

o On the tender stage it is not possible to plan the tasks of different

technical areas, or the need for such planning is not understood”

The problem is related to the fact that the design costs accumulate with the man-hours

spend on the work, and the company is interested in spending as little time as possible

on every task. The lack of time is enforced by tight schedules set by the customers and

the fact that there are many projects that designers have to deal with at the same time.

Therefore, design companies strive to start design work as early as possible, and they

prefer to invest minimum time in planning in the beginning of the projects.

Designing the design is, therefore, an instance of inaccurate and basic project planning

that characterises the industry, according to the team. Planning in general, time-wise

and content-wise, largely relies on project managers, and the input of disciplines is

minimal and inaccurate. All three problems selected by the team relates to planning

the design work and organising the resources. The problems of planning were also

mentioned by the Team 1 as described above, and were broadly discussed by Team 3

on Case X-Y workshops.

Finally, the principles of ideal project outlined by Team 2 included the following

points. Five of these points are the same as suggested earlier by Team 1 (marked with

an asterisk):

“Planning the process beforehand*

Adequate schedule*

Keeping organisation as thin as possible

People work only on this project

People do not change during the project

Finishing phase of the project is maximally short

o Designers start delivering finished work to themselves early enough

Regular internal meetings*

Fast and comprehensive updates when necessary

Organised information exchange

Sufficient resources

People are committed*

62

Owner is skilled and experienced*”

The project manager wished that the organisation in design project was kept as thin as

possible, so that less people are involved in design project and can focus more on one

project. Ideally, the manager said, each designer should work only in one project. The

manager also told that the completion and delivery phase of the projects that usually

takes place after most of the design is done should be kept as short as possible.

4.4. Case projects X and Y

4.4.1. Description

Case project X was a traditional railway renovation project carried out by Company

Beta. The project took 14 month, from spring 2012 to summer 2013, and also included

building a 3D model for machine control systems, contracted at the late stage of the

project with a separate order. Project X was a general design project.

Case project Y was an ongoing railway project in which Company Beta was

participating at the time of research. The project included reconstruction of existing

track and construction of the second parallel track, and the design phase utilised BIM.

The project was carried out by several design companies, and the area of the project

was divided between them in equal parts. One company had already completed the

designing, and the other companies, including Company Beta, had recently started the

project.

Both projects X and Y were carried out by the same project management team – Team

3, which included the project manager and leading designers of major disciplines.

DBB contracts were used in both case projects. In Case project Y, BIM was a part of

the contract for Company Alpha, and was not contracted separately from design. The

designs delivered by all companies were going to be integrated in collaboration model

by Company Alpha.

The two case projects were not comparable because project Y included considerably

larger amount of work (construction of new track) and more resources were dedicated

to the project, since the schedule for the project was set very tight by the customer.

According to the team, the designers in project Y had to focus on the project and spent

most of their weekly work hours on it, whereas in project X the timeframe was longer

and the work was much less intensive, letting designers to focus on other projects.

Project Y also involved much higher number of designers. That explains why, despite

substantially larger amount of work, the project was scheduled for 13 month, against

14 month in project X.

63

BIM in the project Y was required by the customer: the organisation is seeking to

improve its own skills and efficiency. But same applies to the design company as well.

The owner and Company Beta developed and tested modelling guidelines. The main

purpose of BIM guidelines was to make the models usable in subsequent AEC phases,

particularly in construction. The BIM guidelines used in the project had to be always

considered by designers during modelling. However, in all BIM projects clients still

require all traditional 2D documents to be delivered.

As the team said, most of the design software they used in all their projects allowed

3D visualisation, thus, when it comes to the benefits of BIM, the company was mainly

looking for other benefits such as model-based cost estimation. In project X, the

company was ordered to create a model for machine control systems, which is also

attributed to BIM, although the major design project was traditional.

No collaboration model had been created in project Y by the time the workshops were

held. Firstly, by the time of the workshop no finalised designs had been submitted to

make the collaboration model – the disciplines were in the middle of their work.

Secondly, the company was looking for IT solution for collaboration model, as the

access to collaboration models was difficult in earlier projects for one of the offices of

the company due to network limitations.

4.4.2. Workshop discussion

The highlights of the workshops, presented below, will combine discussions that took

place on workshops for the cases X and Y, making necessary distinctions only when

necessary. Presenting the cases X and Y together is required because the projects were

made by the same project team, and hence it is often impossible to attribute statements

made on the workshops to one or another project exclusively. Separating the workshop

data between the cases would also be methodically incorrect, since the designers, when

describing the past events, are prone to a recall bias and might describe each of the

projects through the prism of their general and more recent experience (project Y was

on-going at the time of the workshops, and project X ended a few years before).

Design, modelling and data

Most of the design software in traditional projects provides some 3D view; however,

there are differences between BIM process and traditional process. The team involved

in the cases X and Y has been asked to describe what the role of 3D and 2D designing

in their work is and how these two modes of work change during the design process,

both in BIM and traditional projects.

As the members have told, in all contemporary projects the design software uses some

3D view on the background. In both types of projects, the designers use terrain model

and geometry as the input data, and first work with 3D. Then they move to separate

64

cross-sections for detailed design, and work in 2D to draw the cross-sections and maps.

They use the 3D model to observe the designs. Because 3D view does not allow seeing

the design in detail, the detailed design has to be carried out in 2D.

The main difference between traditional and BIM projects is that in traditional projects

some designs are done entirely in 2D, and final output is always in 2D. The information

is transferred in 2D to disciplines that generally work with 2D (such as bridge and

geotechnical design) even if a designer sending this information used a 3D view in his

work. In BIM projects, there are specific requirements for the model and the process

differs from traditional designing in certain ways. Design is done by modelling always

with 3D output, and 3D data is exchanged and added to the collaboration model.

As the team stated, in the future at least some of the documents could be delivered

only in 3D. According to the team, one of the reasons for why all traditional 2D

documentation is still required is that not all construction companies have BIM

capabilities and could use the models.

In both Case projects X and Y the designing started, similarly to the project B, with

gathering the initial data and evaluating the need for geotechnical research in field.

After geotechnical research was carried out, additional research had to be done again.

In total, ground surveying continued for 6 month. The team also reported problems in

terrain data provided by the client that explained why extra land surveying was needed.

The designing in project Y started with track geometry design; project X was only a

reconstruction project and did not require changes to geometry. As the team described,

in railway projects the track geometry determines overall design to a very large extent,

and is required by all disciplines in order to proceed. In Case Y the geometry design

was done at the same time for all project areas, and all companies could use it as input

data. The companies then had only to match the areas where their designs met.

The railway and geotechnical design usually start at the same time, and typically

ground research is carried out for geotechnical design, as it was done in both Case X

and Case Y. Due to territorial scale of typical railway projects, geotechnical design

plays a very important role in railway engineering. Like in road projects, geotechnical

solutions are required in particular locations of the railway where ground

reinforcements are needed, for example, areas with soft ground. The track designers

draw cross-sections of the railway each 20-40 meters and in specific places where

required, and a longitudinal section that shows the cut of railroad structure along the

track.

As the track design is developed, drainage systems are designed to ensure sufficient

water flux and protection from flooding along the track. Bridges and electric systems

are other large disciplines. As the team described, “every discipline develops more and

65

more accurate design throughout the project, and the designs have to be discussed and

matched continuously, unless the project has a freezing point at which some tasks have

to be completely finished.”

Changes of requirements

In project X the railway had multiple crossings with roads, and some of the crossings

had to be changed to bridges or underpasses in order to increase the road and railway

capacity. The changes in some bridge solution initiated by the customer required

changes in the schedule and additional work order, which made the project longer.

Because such changes are approved and paid by the customer they do not constitute a

problem for the design company.

The change in project X was related to the fact that the bridge structure initially

required by the customer was not possible to be build, because it would have required

the existing road traffic to be substantially cut during the construction. This was

prohibited by the road traffic authority after the designs were sent for inspection in the

middle of the project.

Generally, the problems in customer requirements often cannot be noticed early

enough because the tender documents lack detail, and because the designers do not

have enough time in the beginning of projects to look for all possible inconsistencies.

That is why in case of problems with requirements, customer has to make additional

orders. This may actually be beneficial for a design company but can extend the project

timeframe and does not deliver additional value to the customer.

Cost estimation

In railway projects in general cost estimation plays an important role, especially when

applied to geotechnical solutions and track designs that determine the amount of

excavation work and material required for construction. As the members of the team

said, BIM can help saving construction costs by enabling the comparison of costs of

different options. For example, the designers assess the impact of different ground

elevations and calculate masses to find out the impact on construction costs. In the

case project, models were used for cost estimation. According to the team, the

estimates from the models were more precise than traditional cost estimates.

Cost estimation was done for all design alternatives, always considering all elements

of designs which were affected, similarly to cost estimation in project B. Construction

costs played a critical role in shaping the design; whenever a cheaper alternative was

found it had to be considered. The team stated cost estimation as one of the major

benefits of BIM in their work.

66

Clash detection

Clash detection in railway projects, according to the team, is mostly relevant for

checking the design of tubes, wires and similar objects at the end of the project. Since

project Y did not have a collaboration model by the time when workshops were carried

out, no model-based clash detection was carried out between different technical areas

by that time.

In project X designers had to match their work in a traditional way – with the help of

meetings and communication. As the team have explained, it is a good practice when

the succeeding discipline sends their designs back to those who have provided the

input data, but in the case project this was not done. Clash detection in project X, as

described by the team, was done similarly as in the Case project B.

Planning

The work in AEC firms is seasonal and design projects are usually finished right before

the start of construction season. The deadline for both project X and project Y was

determined by the dates when the construction work had to be put on tender. The

projects were generally scheduled backwards starting from the final deadlines, and this

was reflected in the project schedules. Because Case project X was a renovation

project, it did not require a long period of initial designing. The design disciplines

could start almost simultaneously: track design, geotechnical design and bridge and

structures design.

A difficulty related to planning and management that has been mentioned on the

workshop is inability of managers to track the work progress within disciplines to

know exactly how far the work is. The work progress can only be seen when the

designs or versions of designs are ready and shared. Other designers also cannot see

the progress until the designs are shared. A typical barrier to sharing the designs is that

designers do not want to publish work that is too unfinished from their point of view,

which can be explained by their wish to prevent other designers from taking the risk

of working with unfinished data, but may also have social and psychological rationale.

The difficulty of tracking the work progress and achieving awareness of design means

more coordination has to be done face-to-face. The designers also told it is easier to

track the progress when a large design task is split into smaller tasks.

Because designers normally work with many projects at the same time, one project in

a designer’s portfolio of ongoing projects can be active throughout a long time, but the

designer can spent only several hours a week on this project. This means that the time

schedule only reflects the dates when the task has been started and is supposed to be

finished, but does not reflect the actual amount of work. Looking at a project schedule

67

it is impossible to say how much time the designer works on the project: 4 days a week

or 4 hours a week.

It is not possible to oversee the amount of work that has to be done within a task, or to

say exactly how much of the task have been completed. Therefore, planning is done

according to the experience and project managers have to compromise on the least

important projects and sometimes re-arrange the resources in favour of more important

ones in order to meet the major deadlines. It is important for project management that

designers provide information about their work and give time estimates throughout the

project. This, and the split of large design tasks into smaller tasks mentioned above is

realised in Last Planner system of production control.

In project X, the designers had relatively limited participation in planning, and

planning was carried out similarly to case B. The Case project Y used the first phase

of Last Planner to facilitate coordination and control. The team held meetings every

two weeks to review the schedule. The design tasks, such as track design, were split

into smaller tasks that could be written on post-it notes. Designers themselves have

decided the deadlines for the tasks, according to their estimates on how long it will

take them to complete the tasks. The notes were then put in an order and written down

in a spreadsheet tool.

Nobody could leave the meeting without agreeing on what the schedule is. If someone

could not attend, he was represented by the project manager. In that case the tasks of

this designer were marked with red colour to make everyone see that this hasn’t been

checked.

Data deliveries between designers were reflected as separate tasks, and that is how the

connection between disciplines was reflected in the schedule, for instance, when one

task had to be finished in order for the other one to proceed. Every designer had to

know what they needed from other designers and had to put it on post-it notes. As the

project manager described, it was difficult for the designers to say what data they need

to proceed in their work. The system required designers to change their way of

thinking: they had to understand their own work process and the place of other

designer’s inputs in their work, instead of just being given the data.

On the last planner sessions the team also discussed the ‘just-in-case’ time buffers: the

designers challenged each other to find out if there was any way of making the tasks

shorter, of reducing the suggested timeframes. As the project manager told, the method

gave certain positive results in terms of planning and also helped the designers to

understand each other and interrelations between each other’s work better, although

there were difficulties and the benefits were not very straightforward.

68

Tools

One design software for road design was demonstrated on the workshops. The

functionality included the dynamic 3D view, the possibility to see foundation

inspections and to observe cross-sections in any place of the road. 2D documents could

be produced from the model automatically. The members expressed certain

dissatisfaction with the fact that available software for railway design and other

disciplines did not have the same level of functionality, according to their knowledge.

The designers also mentioned that their work was affected by interoperability

problems and problems with coordinate systems and file formats. The team have

shown a variety of specialised commercial and open-source software that different

disciplines could use in design projects. According to the team, the variety of software

and the lack of interoperability made data exchange between different software

problematic. However, interoperability problems were affecting their work much less

than several years earlier.

Sentiment analysis

Figure 20 presents the results of sentiment analysis for workshops with Team 3. The

sentiment analysis for the cases X and Y was done for two projects together.

Figure 20: The occurrence of topics and sentiments on workshops: Cases X-Y.

The results of the sentiment analysis of the workshops with Team 3 can be placed both

in terms of sentiments and discussed topics between the results of the previous cases.

This is partly related to the fact that in both project X and project Y the team was

0 2 4 6 8 10 12 14 16 18

Collab. Model

Clash/error detection

Cost estimates

Tools

BIM implementation

Modelling /data

People

Collaboration

Project planning

Contracting

Design

Changes, requirements

Sentiment analysis, Team 3

Negative sentiment

Neutral sentiment

Positive sentiment

Number of statements

69

leading the design process, and thus was able to discuss the design, but could also

discuss BIM implementation.

Summary

The discussion of BIM with the team was largely based on the member’s general

knowledge and past experience, since the Case project Y was in the middle of its

realisation at the time the research was conducted. The results of BIM implementation

in project Y, therefore, could not be evaluated.

The workshops revealed the differences between the perceptions of benefits of BIM in

road and railway projects. As the Team 3 stated, clash detection typically did not play

an important role in BIM projects, but the most visible benefit of BIM was cost

estimation that allowed evaluating different options during design and provided more

accurate estimates. In project X a model for machine control systems was created;

another benefit of BIM was possibility to model the space around the track to ensure

that the train can go through without colliding with surrounding structures and

electricity poles.

The discussion largely repeated the discussions of previous workshops, providing

similar findings. The role of the process was discussed extensively, and a number of

problems were described related to planning and resources. Besides the fact that

designers work at one project at a time and the lack of a master schedule, the team

explained how a difficulty to track the design work can affect planning. The role of

designers in planning and coordination of work was emphasised.

The Case Y utilised the Last Planner, which addressed the planning problems

mentioned above. The team held regular planning sessions on which individual

designers communicated their work as broken down in small tasks and provided time

estimates for when these tasks are to be completed. The system allowed better mutual

understanding of design process, but it was difficult for designers to find out what data

they needed to complete the tasks.

4.4.3. General problems in design projects – Team 3

The first problem suggested by the company was lack of communication and

knowledge sharing between technical areas. The team members have mentioned this

issue multiple times throughout the workshop. The disciplines form knowledge silos,

and are focused on their own areas. The problems in communication are enforced by

organisational, technical and physical barriers. Lack of communication negatively

affects the collaborative work.

“Technical areas do not communicate with each other enough

o Interoperability problems

70

o Too many different software

o Manual input of information, both related to software and

interpersonal communication

o Resources change during the project

o Inappropriate project organisation

o Project members work in different offices

o Designers tend to concentrate only on their own areas

o There are knowledge silos

o Project prioritisation

o Line organisation vs. project organisation”

The second problem that was discussed by Team 3 partly repeated the previous

discussion with the Team 2 and related to inability of designers to work only on one

project at a time. Most of the suggested causes repeat the problems discussed on the

workshops with Team 2.

“It is not possible for designers to work only on one project at a time

o No tools for checking the readiness of work of different disciplines

o Lack of cross-checking between different disciplines

o Prioritisation of projects from designers' side (designers, sometimes

unconsciously, prioritise projects that are for some reason more

attractive to them)

o Timeframes for design are short

o Timeframes for bidding are short

o Urgent issues, communication (e.g. emails, phone calls) disrupt the

planned work on a project and force to switch to other project

o Projects that are on different stages (tender, offer calculation, design)

occur at the same time.

o There is no master schedule that would put all project schedules

together”

The last problem related to the work content being not clear in the projects and to the

changes in work content. This makes designing unclear in the beginning and requires

customer to make additional offer. One of the reasons for unclear work content is the

lack of information from the previous design stage. That can be related to low quality

of initial data, or to missing information. Sometimes an entire design phase can be

skipped due to budget or time limitations, and there is not enough information in the

beginning of the following phase as a result. Defining the design intent and work

content might be also difficult if the stakeholders impose conflicting requirements and

have different opinions.

71

“Work content is not clear, changes to work content occur

o Initial data is not available, because it can be e.g. in a wrong format or

coordinates; changing requires a lot of time and effort

o Projects are nowadays faster, design phases are often skipped due to

budget limitations

o There are not enough resources available at the beginning of the

projects

o Not enough initial data of the surroundings, e.g. land use plans

o Difference of opinions between stakeholders

o Schedule

o Customer's budget”

When stating the problems that occur in design projects, the team members mentioned

lack of communication and knowledge sharing between disciplines, problems with

data management and interoperability. The results of the case study with Team 3 were

consistent with other cases and with the concepts identified in the literature review.

5. Discussion and Implications

One of the main goals for the current research was to investigate the current state of

the implementation of BIM and possibly other technologies in infrastructure projects

in Finland, and to explore the nuances of design practice in infrastructure firms. This

was goal was summarised in the third research question:

3. What is the current state of design practice and BIM implementation in the

infrastructure sector in Finland? How do the professionals perceive it?

What is special about the research question is that it contains a sub-question and

reflects the specific way the research was carried out: (1) through collecting general

information about the case projects and (2) through analysing the perception of design

team members of both the case projects and the general practices in the companies.

Because the majority of data was obtained in interview and workshop format, the

perception plays the major role.

The discussion of results in this final chapter starts from answering the research

question three, in section 5.1. The purpose of section is to summarise the results of the

empirical study that have been presented earlier, and the discussion will be conducted

in the following order:

Design process in the case projects

Building information modelling in the case projects

72

After that, the section 5.2 will return to the second research question and suggest how

BIM could look like in the future, and what barriers should be overcome to achieve

more advanced stages of implementing BIM in infrastructure design projects. The

section 5.3 will summarise the findings and discuss some practical implications of this

research with regard to what directions companies can take in order to overcome the

barriers, and section 5.4 will discuss the methodological limitations of the study and

propose some directions for future research.

Finally, the last section (number 5.5) discusses the experience acquired during this

research project, in particular the faced difficulties in collecting quantitative data from

the case projects. This experience may be valuable for researchers who plan to conduct

case studies of design projects in the AEC sector in the future. The section also

discusses how software tools and other methods can be implemented to collect

quantitative data in infrastructure design projects.

5.1. The current state of infrastructure design in the case projects

5.1.1. Design process in the case projects

Concurrent engineering process

The discussions and project diagrams show that the design process in infrastructure

sector is concurrent. The design disciplines work largely in parallel, although

commonly the key disciplines start earlier and their work defines to the large extent

how the design will look and how much it will cost. In the railway projects, such

definitive element of design is track geometry. In road construction the shape of a road,

types of intersections and a number of lanes are defined by traffic planning and

simulation.

In all case projects the capacity was the main priority for the customer and was a 'fixed'

requirement, whereas cost and constructability were variables, aimed to be optimised

during design. This could be related to the fact that in infrastructure projects the

customer is practically always a public organisation, which possesses substantial

resources and can afford certain flexibility in the budget, if spending is justified by the

delivered value. However, the fact that the design is 'fixed' already very early in the

project is not only related to customer requirements. The design is restricted by the

environment: existing infrastructure, surrounding buildings, terrain, ground properties

– all have direct and often definitive impact on design and on construction costs.

The order in which the disciplines develop design is generally defined by their level

of importance to the project requirements, and the involvement of ‘later’ disciplines in

the early design phases is limited. Design in railway and road projects can be

characterised as point-based.

73

Assessment of design alternatives and cost calculation

In these circumstances, the second task of designers after ensuring constructability,

safety and functional requirements, is to minimise construction costs, and to inform

the client of design alternatives when such alternatives exist. Minimisation of costs

requires cost assessment during the project. According to the project teams, in both

BIM and traditional projects costs are calculated for design alternatives when there are

such, and the total construction costs are calculated at the end of every design phase.

Because cost is not the only factor that has to be taken into account, the selection

between design alternatives often requires substantial involvement of the customer.

Cost calculation is performed using specialised software. In traditional projects, this

software uses manual data entry, and calculation of costs requires time and effort.

Therefore, in traditional design there are limitations regarding cost assessment during

projects (e.g. when and how often it can be done). In BIM projects, cost calculations

can be done using data obtained directly from the models, which can speed up the

calculation process and make calculations more precise, as it has been reported by

Team 3. In the future, this can increase the role of cost optimisation in guiding the

designs and result in better optimised projects.

Design integration and cross-checking

Fitting the designs together and detection of clashes is a necessary part of collaborative

design practice in AEC companies. The traditional process of clash detection has been

described in detail by Team 2. Clash detection usually takes place during face-to face

meetings between designers from two or more different disciplines, and large team

meetings. On the team meetings, designers obtain information about the state of work

of other designers, observe each other's designs and coordinate the work. As a designer

from Team 2 mentioned, regular meetings help clash detection because designers get

information on where to look for clashes. The detection of clashes itself usually

happens when designers meet in pairs. These meetings in Case B were very usual and

they were easy because designers worked in the same building. Designers combine

two CAD drawings to see if they fit each other, and discuss what corrections should

be made to designs. Traditional clash detection has limited reliability.

In project A, the detection of clashes was performed by model coordinator or by the

team on review meetings, simply by observing the collaboration model. This manual

approach has limitations. Firstly, it is also prone to human error and does not allow

detection of all clashes. Secondly, it does not allow prioritising the clashes, for

example according to their cost. Thirdly, the time needed to observe the collaboration

model by model coordinator means that the clash detection cannot be done

immediately during design, which implies higher costs for re-design (figure 7, page

22). In the future projects, automatic clash detection can allow faster, more accurate

and reliable resolution of clashes during the design process.

74

Communication

The development of design requires constant coordination effort from involved

designers. All case teams have emphasised the importance of face-to-face

communication and regular team meetings in design projects.

All the case teams had regular meetings in the projects, with different frequencies. The

lack of time can be seen as one of the major barriers to communication: designers

usually have to work on design descriptions very intensively in order to meet the

schedules, and this is usually done individually. Moreover, designers are occupied with

other projects that require meetings too. As a result, there is simply not enough time

for meetings.

None of the teams has been entirely satisfied with communication in their work: at

least some designers in every team have stated that communication between

disciplines was difficult and there could be more communication and more meetings.

This was noticed both in the questionnaire and when making the cause-and-effect

diagrams of problems in design work. The teams have also made it clear that they see

communication as the main facilitator of knowledge sharing, and all teams have

mentioned the lack of knowledge sharing or knowledge silos formed by disciplines

when asked about problems in their companies. Knowledge silos imply the lack of

common language and understanding between technical areas, and the tendency to

focus on own narrow scope of problems during projects: “the technical areas don’t

see what others are doing and do not discuss with each other”.

Design planning and control

Project management was one of the most discussed topics on the workshops. The case

project teams reported difficulties related to project planning and control. Particularly,

the general planning and management practice in infrastructure design firms was

discussed with the teams 2 and 3. Team 3 provided especially valuable information

about control and scheduling in projects and about the team’s experience of using Last

Planner.

The first project management problem to be described here is related to the lack of

visibility of design work in projects. As it was mentioned earlier, designers have to see

each other’s work and conduct face-to-face discussions in order to eliminate

inconsistencies, coordinate the work and to obtain shared understanding of design.

Project managers need to know how far every designer is, if there are any potential

problems or risks for the project and if any actions need to be taken.

The visibility problem results from the fact that design information is not available in

real time as design evolves, and is delivered occasionally by designers, usually when

their tasks are complete or close to completion. As a result, a project manager and the

rest of the team can see the work of every designer only after a rather complete version

75

of this design has been shared. Between the start of the task and the delivery of design

it is hard for others to know how far the work has progressed, which makes

coordination difficult.

Designers have their individual virtual environment in which they carry out their tasks.

They share information with certain regularity, and prefer to share design descriptions

when they are ‘ready enough’. Sometimes, they even tend to ‘hide’ their work on

personal computers, so that unfinished work is not seen by others. On one hand, this

can potentially reduce the risk of making other designers rework in case something

changes later in the shared designs. On the other hand, late delivery of data can cause

unnecessary waiting and will surely make design process less visible to others. Team

2 emphasised that sharing the day’s work on a project drive was considered a good

practice and was mandatory for all team members.

One could imagine that, even if there is no constant possibility to see the work of

designers in progress, it is still possible to estimate how much of the work have been

completed based on comparison between time reserved for the design tasks and time

spent by designers. In Case A, the project manager was able to notice that one of the

designers did excessive work when the amount of hours spent by this designer started

to exceed the plan. This means that the project manager was not able to know about

this problem before it visibly affected the project costs. Team 3 have mentioned that

“For the project manager, the only way (to know about the progress) is to look at the

spent hours. So if the half of the time budget is spent, he can only assume and hope

that half of the work is done.”

Therefore, the second problem relates to controlling the resources in projects: how

much time is being spent on every task. In addition to the lack of visibility, it is hard

to plan how much resources have to be spent and to control it because the outcome is

intangible and the work cannot be calculated in the same way as in manufacturing

activities. Therefore, in design projects it is generally difficult to relate the resources

(time) to the outcomes. This is a conceptual problem that can be seen in other service

activities.

As the case teams explained, it is almost a rule that designers in engineering firms

work on several projects at a time. A traditional project schedule shows the dates when

tasks start and the dates when they should be finished, but the schedule does not show

the workload of designers, or how much time a designer spends on each task.

Designers can spend 4 hours a week on a project or 4 days a week, and this is never

reflected in the project schedules (figure 21).

The case companies do not employ company-wide schedules that could integrate

schedules of all their projects, which makes the division of designers’ time and effort

between projects uncontrollable, and to some extent up to designers’ wish.

76

Figure 21: The design tasks as shown in project schedules; the schedule (on top) is an estimate

and does not reflect the actual amount of hours spent on a project.

According to Team 3, sometimes designers start paying more attention to one project

unconsciously and spend more time on it than on other projects, even if this is not

required. Such ‘prioritisation’ may be caused by creative interest of a designer or his

own understanding of which project is more important. This can have an adverse effect

on other projects.

As Team 3 has mentioned, the ability to control the projects could be increased if

design tasks were split into smaller fragments and reflected in schedule in a detailed

way, and if, as a consequence, design data was delivered more frequently in smaller

batches. This would help designers follow each other’s work more frequently.

Team 3 in Case Y utilised the Last Planner system of production control. The Last

Planner (Ballard, 2000) is a method of planning and control that is based on continuous

planning throughout the project, direct involvement of designers in planning, division

of large tasks into small sub-tasks with clear connections between them, and regular

meetings. The results of implementation of Last Planner in the case project were

positive, although, according to the project manager, designers had difficulties figuring

out what they need from each other in order to complete their own tasks. The method

required designers to change their way of thinking and to learn better how their work

is connected to the work of other team members.

Case Y was an exception: in the rest of the case projects no clear methodology for

planning was implemented whatsoever; the designers had limited involvement in

planning and at most provided time estimates in the beginning of projects.

Coordination and control during the projects, except for project Y, were rather ad-hoc

and the schedules were not detailed, showing only standard high-level tasks that would

typically occupy several months on a timescale. Besides the time component of

planning, the content of work was also not very well defined, according to Team 2,

which sometimes led to changes, additional orders and delays. The schedules were

77

often made only before or at the beginning of the projects and were not updated later

although changes occurred.

Team 2, which was involved in Case B, has stated that the industry does not appreciate

planning as it probably should, and planning is too basic and high-level. The designers

from all the project teams have said that the main reason for the lack of planning in

projects is the time and budget constraints: designers cannot spend too many hours on

planning and too many people cannot be involved in the beginning of the projects,

otherwise the costs would grow. Team 2 has suggested that employing project

management software could be beneficial to aid managers. The team’s statement that

there were no tools, in general, could either mean that the companies were unfamiliar

with the available tools or that there were no tools that could meet the needs of design

projects in infrastructure construction.

Information management

The main information exchanges that occur in projects are deliveries of design data

and meetings between project members. After examining the project swimlane

diagrams and the other data, it has been noticed that meetings, deliveries of data and

start/completion of tasks often occur at the same time, which points at the relations

between tasks and allows attributing information exchange events (data deliveries,

meetings) to the tasks. Since many tasks require long time to be accomplished, the

frequency of information exchanges is not very high in projects. The mutual awareness

of design is generally established on team meetings.

There were also minor problems related to tracking actions and realising what files

carried the newest information. For example, shared project drives do not allow seeing

this information in file metadata. As an option, designer might reflect this information

in file names, although it is not done in every project. Some specialised data

management solutions (e.g. project banks) can track updates and other operations with

files automatically, and record all relevant information in file metadata.

Although in the case projects such solutions were used they were only meant for the

final designs, and internal exchange of design data was done via shared project folders

and even emails. The email was mentioned to be the least efficient channel because of

the limited file size and long waiting times for uploading and downloading, and, most

importantly, because only the recipient of an email could see the file and not the rest

of the team.

In addition, occasional working with incomplete data, as a consequence of parallelism

of design tasks, has been mentioned as one of the problems by Team 2. This problem

was found in the literature, for example, in the article by Eppinger (2001).

78

Summary: process view on design practice in infrastructure projects

The study have not revealed significant differences between the traditional and the

BIM projects in the way projects were managed and work was organised. The

differences concerned primarily the technical aspects of designing. The design

process was characterised by a range of problems which have also been reported in

the literature and which were similar in conventional and BIM projects. The findings

have confirmed the notions found in the literature with regard to the problems of

collaborative design, and have also shown high awareness of the project teams

concerning the problems of their work process.

The list below is a summary of characteristics of projects that were suggested by

designers as current problems (the original lists can be found in each third subsection

of sections 4.2 – 4.4):

- Lack of communication

- Lack of information sharing

- Lack of tools for data management

- Insufficient data from previous phases

- Unclear scope of work

- Tight schedules, lack of time

- Designers work with many projects at the same time

- Project organisation grows too large

- Basic and inaccurate planning

- Lack of tools for project management

- Designers use different formats

- Lack of cross-checking between disciplines

- Long finishing phase of design projects

In general, the ad-hoc project planning, lack of work control and coordination were

the most discussed topics related to design. The lack of proper information

management was also a much discussed problem. BIM and other technologies and

methods (project bank solutions, project scheduling and control software, Last

Planner and possibly Big Room) can be used to reduce the impact of some of the

discovered problems, for instance, the low visibility of design outside the discipline

creating it.

Generally, the need for process improvement is recognised by companies but the

resources are scarce to reengineer the processes and designers are used to old

practices, which makes any change difficult. How this affects the implementation of

BIM and what the potential of BIM to improve the design is will be discussed in the

following sections.

79

5.1.2. Building Information Modelling in the case projects

Perception of Building Information Modelling by the case teams

When the members of the design teams were asked to state the characteristics of BIM

that they knew in an open-ended question, the team members have stated a variety of

benefits, most of which are also found in the reviewed literature. The suggested

benefits are presented on figure 22, and have been shown in more detail on figure 16,

section 4.1., page 42. Among the benefits suggested by the researchers in questionnaire

and in discussion, the most prominent were: (1) design data integration and clash

detection, and (2) cost calculation and evaluation of design alternatives.

As Team 3 reported, cost estimation was done with better precision with the help of

BIM. Cost estimation was seen by the company as one of the primary benefits of BIM

in railway projects. In the Case A, which was a road design project, the manager

viewed model-based cost estimation during design as one of the potential benefits of

BIM, although in the case itself cost estimation was done in a traditional way.

However, according to the case team, BIM was used for decision making on the initial

design stage primary to the project A, for evaluating various deign alternatives with

the customer. In both BIM cases the models were used to present the designs to the

client and public.

Figure 22: Open-ended question about the features of BIM – answers categorised and

arranged in a stacked bar chart.

0 2 4 6 8 10

Error/clash detection

Combining designs of different disciplines…

Data utilisation through lifecycle, data reuse

Process/work efficiency

Information exchange and transfer

Mass calculation

Easier change implementation

Machine control, prefabrication

Visualisation, 3D

Keeping data in one place, data administration

Reduction of unnecessary work, re-design

Automated creation of drawings

Collaboration and communication

Earlier utilisation of initial data

Increased amount of work

According to your understanding, what are the most important features of BIM?

80

The role of collaboration model for clash detection was especially important in Case

A, which was a road project. In this case it allowed detection of two clashes that would

not have been recognised otherwise and that constituted a substantial risk for

construction. Team 1 stated that, according to their client, the possible savings from

detecting these clashes alone covered the whole cost of BIM project. Team 3, which

was involved in railway projects, stated that clash detection was important in particular

cases, but it was not as important benefit for railway design as, for example, cost

estimation. Overall, clash detection was the most recognised benefit of BIM among

the participants of the research, according to the questionnaire study.

In spite of the benefits, it is unclear how BIM-based clash detection affected the design

integration in the Case project A compared to traditional projects: whether it was

performed in addition to traditional clash detection process, or whether it to any extent

replaced the traditional process, making the designers rely more on the instructions of

model coordinator than on peer-to-peer meetings. If the latter was the case, it could

have had an adverse effect on communication by decreasing the information exchange

between disciplines. The measurement of impact of BIM in its current state of

implementation on communication in projects requires in-depth analysis that is beyond

the possibilities of this research, but it is an interesting topic for future studies.

In addition to implementing BIM in projects A and Y, digital models for machine

control systems were made in project X after design was finished. With regard to other

benefits suggested by designers, there is less strong evidence that they were realised

in the case projects, since most of them are very intangible.

The way how Building Information Modelling was perceived by the case teams is

characterised by a contradiction. In the questionnaire, the benefits of BIM were rated

differently, depending on how questions were formulated: whether designers were

asked about the general properties of BIM or about their experience of BIM in projects.

For example, the parametric object-based modelling can make some routine design

tasks easier to perform (Eastman et al., 2008: 14), and this can explain why such

benefits as work efficiency and easier change implementation were suggested by

designers (see figure 22). However, when asked to rate a pre-defined set of impacts of

BIM on the actual projects, the designers have given negative ratings to several

process-related characteristics: the impact of BIM on project delivery time, time

required for changes and additions and ease and joy of working (although the latter

caused controversy among respondents). The fact that these characteristics were rated

negatively directly contradicts the responses shown in figure 22.

The actual impact of BIM on parameters related to information exchange and

management is also unclear. Although designers have stated that the impact of BIM is

positive, the overview of current design process suggests that the information

81

exchange in BIM projects is identical to such in traditional projects. The fact that

model integration and a major share of modelling were performed as a separate process

by BIM experts, suggests that the information flow in BIM projects could be somewhat

more complicated than in conventional projects, especially since all traditional

documentation in such projects is still delivered to the customer.

When asked to rate the level of satisfaction with the process of working with BIM,

designers, who have previously been involved in BIM projects, gave very low ratings

relative to the scale: 4,5 out of 7 by experienced BIM users and 2,75 out of 7 by less

experienced users (see figure 13 on page 39, section 4.1). On the other hand, when

asked about the willingness to work with BIM in future, the designers gave high

grades: 6,83 by experienced users and 5,75 by less experienced ones, on the scale of 7

(see figure 12 on page 39, section 4.1).

During the workshops, the members of BIM projects were especially critical of many

process-related aspects of implementing BIM. The low quality of models provided

by third parties and by designers in the teams, the difficulties in exchanging

information and in communicating instructions to designers, the lack of standards

and guidelines – all these problems were mentioned very often. The critical attitude

towards the process is especially evident from the sentiment analysis charts presented

in sections 4.2-4.4. Some professionals on the workshops, while acknowledging the

benefits of BIM for the owners and construction phase, referred to it as ‘additional

work’ on design phase.

These results have two important implications. The first one is that the problems that

occur when implementing BIM are largely related to process, as it has been initially

thought when planning this research. This means that the potential of BIM is not

realised to the full extent and the barriers to it should be overcome by industry.

The second implication that results from the controversies in the responses is that

there is a difference between how designers perceive the potential benefits of BIM

(what BIM is supposed to give to designers) and how they perceive the actual design

and modelling process in their companies (what BIM actually gives to them).

Although BIM already provides significant benefits, such as 3-D visualisation of

designs, timely detection of errors and clashes and the ability to choose the best

alternative using automatic cost estimates, these quality-related benefits are not visible

for the individual designers working in projects, since many of them materialise only

during or after the construction phase. The use of BIM was still mostly the initiative

of project owners, although the case companies also valued the experience they

acquired during the projects, and recognised this experience as their competitive

advantage. As the project manager of Team 3 has mentioned in the interview, “all the

new things can have a negative effect, but we (the case company) understand that this

82

(BIM) is future, you can’t do the design always in a traditional way. If you want to do

design in future, you have to learn BIM.”

The merit of calculating ROI of BIM implementation in design projects.

None of the case projects attempted to measure the return of investment in BIM

implementation or the economic effect of any of the benefits of BIM in particular. As

described in the beginning of this study, conducting such measurement in the context

of intangible project-based activity is a very hard task. In addition, the companies do

not have resources, especially time, to conduct extensive studies. For design

companies, the only indication of whether BIM is beneficial or not is the possible

additional profit in a BIM project. It is difficult to attribute a particular difference in

design costs between projects to the fact that BIM was used in one of them, and

separate this from other factors, or to measure the improvement in design quality.

The discussions with the teams and the study of perception suggest that the companies

rely on their vision of the future when developing and implementing a new technology.

Measuring the benefits of BIM is not only difficult because of the lack of instruments

and resources for data collection. With regard to the design phase of infrastructure

projects, it is difficult because there may not be many benefits for designers on the

current level of development of BIM at all, although there are benefits for the

construction phase. Nevertheless, the additional revenue received by design firms in

BIM projects can overweigh the possible costs of additional work. Finally, a set of

innovative technologies and approaches like BIM is not a product of a laboratory – the

only way it can be developed is through years of application in industry.

Even the professionals cannot say exactly in what direction and how far the technology

develops in several years. The fact that managers cannot calculate exactly how much

money the technology will return does not prevent more and more companies to join

the BIM race, even though the technology could be not very rewarding for them in a

short term. The likely motivator for companies is not only their perception of current

and future benefits or the willingness of their clients, but also the pressure from

competitors, who may “get there first”.

It can be concluded that the need for quantitative studies of benefits of BIM for design

at this level of development is partially overestimated. However, there is a certain risk

that the industry may stop the investment in technology if the benefits are not proven

with time. The measurement in design phase is needed as a part of development of

BIM, in order to benchmark the different ways of implementing BIM in design projects

and find the best practices.

83

The deficiencies in the current state of BIM process

BIM, as it was implemented in the case projects, gave little benefits to collaboration,

as in neither of the two projects that utilised BIM could the designers use the

collaboration models.

According to the description of modelling process by the project A team (presented in

subsection 4.2.2), there was a substantial disconnection between design and modelling

in the project. One of the reasons for that was that the designing was already taken to

detailed level on the previous design phase: most of the alternatives have been already

evaluated and selected. Therefore, the primary function of BIM in the project was clash

detection.

The model coordinator was responsible for the integration of all design data. The data

provided by designers in most cases needed some modification before it could be

included in the collaboration model, and this modification required manual inputs. The

swimlane diagram of the project created together with the team shows that this

integration and correction process took around three months.

The correction and completion phase can be viewed as a bottleneck in the project

workflow, at least in Case A. The process relied mostly on model coordinator, and,

according to largely negative sentiments on the workshops, could place excessive

burden on him and other BIM experts. Model integration in the project was done at

the later phase of designing; there was a substantial time lag between receiving the

designs and addition of data to collaboration model.

This type of BIM process is illustrated on Figure 23, a diagram based on the swimlane

diagrams of the case projects created during the study. The figure shows that the

process of correction and completion in BIM projects is performed as a cycle, and

starts at a late phase of a project after most of solutions are ready. The diagram also

shows data exchange practices and tools.

In the Case Y, the collaboration model was not created by the time of the workshop,

five months since the start of the project, which suggests that the characteristics of

BIM process were not much different from project A.

The deficiencies of the current BIM process, the various problems that occur when

implementing BIM, and the lack of visible benefits for designers can become objective

reasons for designers’ resistance to the technology, even though the impact of the lack

of knowledge and natural resistance to change should not be discarded. As the

questionnaire has shown, the designers have negatively rated the impact of BIM on

project duration, duration of tasks and ease and joy of working. Moreover, the number

of negative statements is higher among the designers who have worked in just one or

84

two BIM projects. This shows that there is some resistance to new technology and lack

of knowledge of benefits among designers with little BIM experience.

Figure 23: BIM process in the case projects, stage 1 of BIM implementation.

The results of the questionnaire show that novel users who experienced only 1-2 of

BIM projects have less precise understanding of the benefits and objectives of BIM

compared to BIM experts. Novices tend to notice the somewhat negative impact of

BIM on their work more than the experts do, since they naturally have difficulties

learning and adapting to new methods of work. Finally, novices have less satisfaction

in BIM projects.

The absence of BIM users with experience of 3 to 6 projects can suggest that the

transition of designers from novice to expert level with regard to BIM knowledge is

currently unachieved. The low number of BIM projects and the temporary nature of

project teams can be a barrier for systematic increase in BIM experience among the

BIM users.

The fact that the teams break apart and the knowledge flow disrupts after each project

has a negative impact on knowledge dissemination, and this problem is typical for

project-based business (Gann and Salter, 1998). The problem can have a major impact

on implementation of BIM. If a designer is taken to one BIM project only and does

not continue working in projects that use BIM afterwards, he is not likely to receive

the amount of skills and knowledge that could give the actual advantage to his

company. The significant gap between the BIM experts and new BIM users, which

85

manifested itself in the questionnaire, might point out at areas where knowledge

management could be improved in companies.

Team 1 has expressed their dissatisfaction with designers’ response to BIM

particularly often during the workshops. The BIM experts agreed that the common

designers in the project followed instructions at most. There were problems with both

the designers’ attitude and the quality of supplied models. Taking that into account, it

can be stated that the gap between new BIM users and BIM experts reflects the gap

between BIM process and designing that is depicted on figure 23.

The overall state of BIM implementation in the case projects could be placed between

stage 1 (object-based modelling) and stage 2 (model-based collaboration) of Succar’s

BIM maturity model (2009). The case companies employed various modelling tools

within several disciplines, and there was transition of models between different

companies and project phases. However, such transition was largely jeopardised by

the lack of software interoperability and various problems with model quality, as in

Case A. Complementary data transfers and manual modifications to the models were

needed in many cases, both in data transfer within companies and between different

project parties. Therefore, although the state of BIM in the case projects was somewhat

beyond just object-based modelling, it still could not meet the criteria of model-based

collaboration.

Summary: the level of BIM maturity and identified problems

The level of implementation of BIM in the case companies is mostly consistent with

the first stage of Succar’s BIM maturity model (2009), and only some parameters as

utilisation of BIM between design phases qualify for the second stage. The current

BIM process in one design phase and one company, as depicted on figure 20, will

be considered in this research a first stage of BIM implementation in a design

company. On this stage, BIM does not have any major effect on collaboration and

information exchange; the designers do not use the collaboration model.

The correction and completion phase of the project is affected by BIM on this stage,

since clash and error detection is used to identify and correct errors in design. This

process may add extra design iterations and require additional time at the end of the

project, but this is beneficial for the quality of design and reduces the clashes. In

addition, the use of model-based cost estimation results in faster and more precise

estimates. The value of utilising BIM in the projects is actualised mainly by savings

of construction costs for the owner. On the current level of implementation of BIM,

the designers cannot sense the benefits of the process.

Here are the list of problems related to implementation of BIM that were formulated

by the Case A and Y teams:

Collaboration model is under-used

86

Faulty initial data

Lack of quality checking

Lack of process guidelines

Interoperability problems

Too many different software products

Manual input of information is required

Lack of software skills

Designers use different formats/templates

Updates to designs not timely enough

Old habits and change resistance

Design and modelling is separate

The major problem identified in the case projects is the low quality of received and

created design data, which includes the initial data, the designs created internally

and the data received from third parties. The quality of data does not meet the

standards necessary for seamless model integration, and most often requires some

intervention from the model coordinator (integrator). The quality problem results

from the lack of modelling skills among the participants, the lack of proper internal

quality checking and the lack of guidelines for modelling and quality checking.

The fact that the process quality does not satisfy the teams can contribute to the

decrease in motivation and the resistance to change among designers with regard to

BIM implementation. Some of the results indicate that there is no systematic

organisational learning policy in place to allow employees sustainably achieve high

levels of BIM expertise. In addition to the process-related problems, the companies

lack the knowledge of available BIM tools and the expertise in using them; they also

lack the expertise of creating the proper software and network infrastructure for the

projects that would allow interoperability and fast exchange of information. There

could as well be shortcomings in the software tools, especially in the older case

projects.

5.2. Towards BIM-based collaborative design in construction

This section returns to the potential role of technology in infrastructure design practice,

which was previously discussed in the literature review chapter. The section mainly

discusses the achievement of collaborative BIM process, which was not in place in the

case projects. The section proposes some additional answers to the research question

2, now taking the results of the empirical study into account:

2. What is the potential role of BIM and other technologies in maximising the

efficiency and effectiveness of collaborative design in construction projects?

87

5.2.1. BIM as a tool for collaborative design

The traditional design process reminds of a painting workshop in which a number of

artists work on one painting. The difficulty is that the specialised artists have to work

in different rooms on their own fragments without the possibility to see the whole

painting. Their fragments, however, should perfectly fit each other, and therefore the

artists have to meet regularly to discuss and exchange their work. The artists should

also be very fast and efficient in order to finish every painting in time, although they

have many paintings that they are simultaneously working on, sometimes switching

from one to another few times a day. This absurd example may illustrate where the

problems of collaborative engineering in such industries as construction stem from.

Building information modelling can make it possible for building and infrastructure

designers to see both their own part and the whole design when doing their work. BIM

could resolve the issue of design visibility by providing a shared virtual 3D

environment, in which all designs will be integrated from the very beginning of the

project, ideally – in real time.

In BIM projects, designers can work in 2D or 3D on their own fragment of the work

and these discipline-specific models have to be integrated into a collaboration model.

In the case projects, as shown on the process model on figure 20, this integration

process was carried out at the later phase, mainly by the model coordinator (therefore

this role is also often called ‘model integrator’). The purpose of BIM, as it has been

already discussed, was to detect clashes in all projects and estimate the project costs

in the railway project; the models were also seen to have value for the customer as

such. The value of BIM implementation, therefore, is likely to actualise mainly on the

construction stage of the projects. The collaboration model was not used by designers.

Realisation of benefits of BIM during design and for design requires substantial change

in the design companies’ understanding of the goals of BIM and its role in the design

process. A number of articles reviewed during this research explore the benefits of

BIM server technology and IFC-based data exchange for design collaboration. These

studies emphasise the potential role of BIM in future projects as design data repository,

which can largely replace the conventional data exchange process and serve as a

multy-disciplinary collaboration platform. (Singh, Gu and Wang, 2011; Poerschke et

al., 2010; Plume and Mitchell, 2007)

One of the experiments when such BIM-based collaboration was achieved is described

the study conducted by Poerschke et al. (2010). In the research, the university students

of three architecture and engineering departments of 3rd to 5th year of studies worked

in a collaborative studio that utilised BIM server technology. The students used BIM

for data collection, analysis, design development, data coordination and project

presentations.

88

As the study reports, “Students from different disciplines could draw upon one central

model, redefine it for their own needs, perform complex analyses in discipline-specific

software using separate models, and then re-inform the central model (with some

hurdles of information backflow to the model). In contrast, a studio that embraces just

one discipline seems limited in exploring the collaboration potentials of BIM.”

However, the difficulties faced in using the collaboration model during the experiment

included interoperability problems, data administration, lack of BIM skills and

education, and a need for detailed workflow planning, which is similar to the results

of the current research. (Poerschke et al., 2010)

The potential benefits of using BIM as a cross-disciplinary collaboration platform are

especially important in the light of the problems of traditional design process in the

AEC sector found in the literature and revealed in the empirical part of this study (see

subsection 5.1.1). Firstly, the use of BIM collaboration servers and the project data

handling solutions (project banks) can resolve such problems as versioning and

availability of data. Secondly, the problems related to project-management and

workflow planning, such as undetailed scheduling and difficulty to see the progress of

work in individual disciplines, are likely to be resolved with the use of collaborative

BIM. What is particularly important is the interrelation between the use of BIM as a

collaboration platform and the reduction of batch (increment) size in the design

process, since the collaborative BIM requires more frequent sharing of design data.

As discussed earlier, the division of general tasks into smaller sub-tasks and therefore

the more detailed reflection of design process in schedules, possibly using the Last

Planner system, can benefit project management. The more frequent delivery of design

data can help collaboration by enabling early cross-checking and analysis of designs.

As the reviewed literature suggests, smaller and more frequent design iterations or

increments can also reduce the waste related to design changes, similarly to the concept

of Agile (section 2.4). At the same time, the frequent uploads of design data to

collaboration model starting from the earliest weeks of design project, and the

opportunity for designers to use collaboration model, are the main characteristics that

make the difference between the current process identified in the case projects and the

collaborative BIM process.

The next step in developing the BIM process is illustrated on Figure 24. The

illustration shows differences from the current process depicted earlier on Figure 23

(subsection 5.1.2). There is no division between the design phase and the

integration/correction phase any more: the additions to collaboration models and

model revisions are done frequently throughout the project.

89

Figure 24: Potential stage 2 of BIM implementation: revisions are done frequently during

design, BIM is used as a collaboration platform.

The letter N on the diagram represents the number of increments made to the

collaboration model; the number of increments should be reasonably high and should

be chosen based on how much time is required to integrate the discipline-specific

design data. When the process, skills and technology improve, this number is possible

to increase.

Possibly, the future technology would allow saving designs directly to a network-

based collaboration model and would not require any model integration and updates

at all. On this level designers would be able to see both their discipline-specific data

and the collaboration model in real time while their work proceeds. This possible

future BIM process is depicted on figure 25.

The process models depicted above are related to the capability stages of Succar's BIM

maturity model (2009). The Succar's capability stages describe the steps towards

collaborative BIM process in the whole project between different stages of life cycle

and/or companies, but also mention the level of BIM-based collaboration between

different disciplines which can also take place within one company. The models

proposed in this work describe the level of maturity of BIM mainly within one

company and one design phase. Nevertheless, the process models above illustrate the

similar idea as the Succar’s work. In both models the level of collaboration is

dependent on the technology, for example, the advanced network-based (server)

technology is also mentioned in Succar’s model as a pre-requisite of achieving the

Stage 3 of BIM capability.

90

Figure 25: Potential stage 3 of BIM implementation: shared virtual design environment,

automatic integration of design data and clash detection.

5.2.2. Barriers for achieving BIM-based collaboration in design projects

The existing time gap between designing and the integration of discipline-specific data

in a collaboration model makes the model not very usable for designers. To make the

model useful for them, the integration lead-time should be as short as possible, and the

fragments of design should be integrated starting from the beginning of projects, so

that the team can see and use the collaboration model right after they create new design

data.

The additions to collaboration model should be made more frequently (figure 24), and

in the best case – immediately after new data is delivered (figure 25). This will also

allow performing clash detection and cost estimation earlier. Since the cost of re-

designing increases when design proceeds (figure 7, section 2.4), making clash

detection as early as possible will contribute to the decrease in rework needed to fix

the clashes.

According to the case data, the integration of models is the key technical barrier in

projects to achieving this stage of integration of BIM into the design process. On this

level of technology, process and skills, the individual discipline-specific designs

cannot be taken to collaboration model so easily. As the designers from Team 3 have

said, “Ideally, we should design by modelling”; but this is not achieved currently.

91

As the Case A have shown, there were substantial problems with the quality of data

provided by third parties, including the initial data, and with the data provided by some

designers within the project team. The model coordinator had either to fix most of the

models when the required corrections were less substantial, or send the models back

to designers. This cycle of correction has repeated several times. The difficulty to

transfer design data to collaboration model can be explained by the following

problems, which have all been mentioned on the workshops:

1. Traditional 2D-based designing is still used by some disciplines, and in those

cases modelling has to be done separately by a BIM specialist after the design.

2. Low IT skills of designers, little or no experience of working with BIM.

3. Lack of modelling guidelines, standards and formats (e.g. designers use

ambiguous coordinate systems).

4. Lack of quality checking procedures for data prior to submission, both for

designers and for third parties.

5. Poor software interoperability, although this has been significantly improved

in the recent years according to the case teams.

6. Suitable software is lacking or not used for some of the disciplines and

applications. For instance, in the project A the bridge designer has reported

difficulties because there was no modelling software completely suitable for

complex bridge geometry. In the same project, the poor quality of models made

by the third party was assumed to be caused by the use of unsuitable software.

7. Required IT infrastructure is lacking in some companies for the use of shared

collaboration models. For instance, in Case Y no suitable technical solution for

collaboration model server have been found by the middle of the project.

8. Outdated BIM tools: in some older software 3D models are displayed in a

separate program and it is technically impossible to observe 3D views while

designing; in the newest software this has been improved.

9. Difficult navigation in 3D models, especially for unexperienced designers.

While the technology-related barriers cannot be overcome by designers and are the

primary concern of software developers, the design companies can improve their BIM

skills and processes. In the advanced BIM process of today it should already be

possible to reduce the modelling lead-times and achieve regular model integration at

least for the major disciplines.

5.3. Summary and practical recommendations

In this section the implications of the findings for design companies are discussed.

Possible directions for future research are proposed, taking into account the experience

obtained when conducting the study. The section aims at answering the last of the four

92

research questions, not definitively but through argumentation and suggestion because

the proposals have not been practically tested in this study:

RQ4. What actions can be taken by the design companies in infrastructure sector

regarding the technological and process development in order to increase

performance in design activities?

The currently achieved first stage of BIM implementation, which has been described

in detail in subsection 5.2.1, means that the potential of BIM to serve as a tool for

collaborative design and improve the design process is not realised. In the case

projects, none of the designers used the collaboration models and the creation of

collaboration models by model coordinators was done or was planned to be done at

the projects’ later phase.

The detailed study of case project A has provided evidence that the way the modelling

process is organised in the case project can extend its completion phase. The

responsibility of a model coordinator is to control and improve the quality of data

provided by specialty designers and to provide them with instructions for correcting

the designs. This lengthy process can possibly slow down the work at the later stage

of a project, making model integration and correction a bottleneck.

Such benefits of BIM as clash detection and cost estimation, however, have a decisive

impact on projects through improving the design quality. This impact is realised

mostly by the customer, for instance, in the Case Project A the client has reported

major savings from clash detection. The companies are currently moving towards a

higher level of integration of BIM in the design process.

The barriers towards achieving collaborative BIM include the different factors that

force designers to use traditional methods and make the integration of designs into a

collaborative model difficult. The large share of designs is still done in 2D and the

quality of provided data undermines the ability to integrate it. There is generally a lack

of process guidelines and standards for modelling and exchange of information,

although the degree of similarity between projects of one type in infrastructure sector

(e.g. road projects, railway projects) provides the possibility for standardisation in the

future with adjustments for the needs of disciplines, similarly to the way traditional

design was standardised in the sector.

The capability of design companies to implement BIM more effectively depends

largely on the skills and attitudes of employees. The questionnaire study revealed a

gap between common designers, who are in most cases novel users of BIM, and BIM

experts. This gap manifests itself in differing levels of knowledge in BIM-related

topics and differing attitudes to BIM and its impact on work process. The attitudes of

novel users tend to be more negative. At the same time, the study shows no respondents

93

who would be in transition between a novel user and an expert: there were no

respondents in the case teams with the experience of 3-6 BIM projects, and the only

identified groups were novel users (1-2 projects) and BIM experts (>7 projects).

Such obvious division in the levels of experience may be an indication of a disruption

in learning. Designers from both companies have mentioned that teams in their

companies were usually temporary. Even if the management team stays unchanged

from project to project, a large part of design team is different every time. This means

that there is a high chance that a designer participates only in one BIM project and

after that does not participate in any BIM projects for a long time. It is likely then that

a company will get many of its designers to participate in one project that uses BIM,

but that practice is not beneficial because one project does not give a designer the level

of knowledge and skill that could make a difference for the company.

This claim is supported by the results of the questionnaire study and the opinions of

BIM experts. Disruption in the learning process at the end of a project is typical for

project-based firms in general (Gann and Salter, 1998). Unchanging BIM teams,

systematic mentoring practices and emphasis on knowledge exchange have a potential

to enable the transition of employees from novice level to BIM expert level and can

also improve the way designers work together in teams.

The study of design process in general have revealed several areas where improvement

can result in leaner, more efficient designing, regardless of whether BIM is used or

not. While the previously mentioned areas of improvement relate to overcoming the

barriers to collaborative BIM, the following areas relate directly to the problems of

designing (figure 26). The detailed description of these problems have been presented

earlier in subsection 5.1.1.

Figure 26: Focus areas for design companies in infrastructure sector

Collaborative

BIM

Efficient and

better design Design process:

current stage

BIM: current

stage

Actions:

overcoming

barriers to

collaborative

BIM

Actions:

providing

better support

for design

process

Measuring

the benefits

of BIM

94

The first area is information management: the project teams have reported some

difficulties regarding the management of design data, management of updates,

tracking of actions with files and data availability. When the necessary data is

unavailable for a user it can result in a delay or force this designer to work with

incomplete data. A designer can also work with outdated or incorrect data. The

suggestions of the project teams have fallen into categories of problems proposed by

the reviewed articles on lean information management. The problems can be addressed

with the introduction of advanced information management tools, e.g. project banks.

The other area not directly related to BIM but more global from the perspective of the

projects is project management and control. The design companies are restricted by

the very nature of the project-based business, service operations and tender

contracting, which all introduce a large share of unpredictability to companies’

operations. As a result, the companies have to manage and allocate the resources

between the projects ad-hoc, in such a way that most important projects are prioritised.

The general lack of time in projects results in simplistic planning practices (e.g. lack

of master schedules), inaccurate scheduling and too general planning of work content.

The simplistic scheduling practice and the inability to track the work progress of

individual disciplines and designers, which were reported in the case projects, do not

help overcoming the unpredictability. This results in the situations when the projects

or tasks have to be compromised in favour of more important ones, the tasks have to

be transferred to designers that are less occupied and the project organisation has to be

extended.

From that point of view such approaches to systematisation of workflow and control

as the Last Planner System of Production Control, can give large benefits to design

teams. The mentioned system has been implemented in Case project Y and, according

to the project manager, gave generally positive results, improving both the

manageability of the collaborative design process and mutual understanding of the

work process, although it was difficult for designers to adapt to the new method

(subsection 5.1.1). The concepts of managing workflow through understanding task

dependencies and splitting large project goals into smaller manageable tasks, are

suitable for the collaborative BIM process suggested in section 5.3., which seeks to

deliver smaller batches of data throughout the design projects. To support project

management, software tools for management, production scheduling and control can

be utilised by the companies.

Here is a short summary of actions that can be employed by the design companies:

Moving towards integration of BIM in design process

o (“Ideally, we should design by modelling”)

Developing guidelines for designers

95

o Developing modelling guidlines

o Standardising model quality checking, formats, coordinates, versioning

etc.

Improving BIM expertise

o Involving the same team in many BIM projects, systematic learning

o Achieving high level of BIM expertise for designers involved in BIM

projects

Employing data management systems for information exchange

Improving project management practices

o Splitting major tasks into smaller subtasks

o Involving designers in planning and coordination

o Improving workflow coordination: e.g. Last Planner system

o Using software tools for project management

5.4. Validation of the results and recommendations for future research

This study is based on the empirical evidence from the case projects and the

questionnaire conducted with the case project teams. This approach, while providing

in-depth exploratory view of the research subject, has limitations regarding the validity

and credibility of results, therefore at least some of the results of this study should be

tested and inspected in a better detail in future studies. The same applies to hypotheses

that are based on theoretical argumentation but are not broadly tested in this study.

Both case projects that used BIM incorporated the second, general phase of design. It

can be interesting to study the implementation of BIM in other design phases, because

the importance of the different benefits of BIM and the process of using BIM may

differ between design stages. For example, the Case A team has mentioned that BIM

was used on the initial stage for evaluating alternatives, while on the general stage the

solutions were already ‘fixed’. This may explain why the main benefit of BIM in Case

A was clash detection. Cost estimation can play a larger role on the initial stage. The

implementation of BIM on the initial and the construction-level stages should be

studied in future.

One of the observations made during the research that requires deeper analysis is that

the current level of implementation of BIM in design companies may have adverse

effect on the timeframe of the design projects. It is possible that in the case projects

BIM can result in a bottleneck at the later phase, since neither the technology nor the

skills of designers allow fast integration of designs to collaboration model, and because

the integration and detection of clashes typically rely on one or two persons. This can

also hamper communication between designers, since in traditional process all cross-

checking and correction activities are done jointly by designers, on face-to-face

meetings.

96

The evidence that BIM on the level observed in the projects can make projects longer

is found both in the swimlane diagrams (long integration and correction phases in

projects) and in the questionnaire results (designers have indicated that BIM can make

both design corrections and the project in general take more time). A study of a higher

number of BIM projects is needed to look more deeply into this problem. At the same

time, the process development actions by companies aimed at deeper integration of

BIM into the design process discussed in this study might help to mitigate the possible

problem.

Another potential topic for future studies relates to perception of BIM. The participants

tended to mention various problems related to BIM implementation and

interoperability as the current problems, however, stating later that some of these

problems have been mostly solved. A large number of problems related to usability of

BIM software and certain software functions, resolved in the past, have been also

mentioned by designers. This leads to a hypothesis that negative perceptions remain

for some time after the problem is solved, in other words there is a lag between the

improvement in BIM and the change of perception. This hypothesis can be tested in

the future studies.

In general, the current study is based on only four case projects (two BIM projects) in

two Finnish companies and may not represent the average state of the industry to the

point of time when the research was conducted, especially when talking about the state

of industry in other regions. Therefore, there have been limitations in answering the

third research question about the current state of BIM and design practice in the AEC

sector. Using methodologies other than the case studies and covering larger groups

would allow validating the distinct findings made in this research and generalising

findings for larger groups with higher level of certainty.

In order to establish more generalised results, questionnaire studies can be conducted

similar to the one in the current research project but with higher number of respondents

and companies involved. In this study the questionnaires proved to be very beneficial

for understanding how professionals perceive the design process, its inherent problems

and the impact of BIM, to assess their level of knowledge of BIM and their

expectations for better design process and technology.

The triangulation achieved in this study through the use of different types of data and

different methods of data collection and analysis provides sufficient credibility of

results within the defined scope of the study. The results of the empirical study are

supported by the reviewed literature and follow a line of theoretical argumentation.

The future effect of BIM and technology in general on collaborative design in AEC

sector needs to be studied more extensively. The needs of industrial collaborative

design include activities that are not attributed to designing as an individual task, for

97

example communication, coordination, planning, control, information handling.

Although ICT has provided designers with drawing and engineering tools and

information exchange channels, much of the information processes are still paper-

based, and management and coordination processes are very basic and are not

supported by specialised tools. Although much of the research is focused on

construction activities, more efforts are needed in the research on AEC design

activities and development of tools and methods for designers.

One of the hypotheses that occurred in this study and was later found in the literature

was that designers operate in two basic modes of collaboration:

close collaboration in a shared design domain, e.g. on a meeting, and

individual designing, in which contributions are done to areas of expertise, to

a large degree independently from other designers.

This observation is mentioned in a study on collaborative design by Kvan (2000). The

author describes closely-coupled design, when designers continuously observe and

understand each other’s moves. At any stage the observer cannot identify a discrete

contribution to the design product from one participant or the other. However, Kvan

adds that much of the designing is actually loosely-coupled, with each participant

contributing to the design within his sphere of expertise. In this case, the experts

operate in their own domains on a shared problem, the design moves in discrete steps

that can be observed and distinguished from other designers’ steps (Kvan, 2000).

Potentially, collaborative BIM and co-location can blur the borders between individual

and group modes of designing.

The other hypothetical reason why independent, loosely-coupled designing is

necessary is that working individually is possibly psychologically easier and could

allow better concentration and productivity when more enduring tasks are concerned;

this aspect of co-working is, however, out of the scope of this study. There are attempts

in other industries to bring two or more designers to perform continuous design activity

together, for example, peer programming.

5.5. Exploring the possibility for quantitative research on AEC design

activities

The initial goal of this study was to employ quantitative methodology to measure the

benefits of BIM and process improvement in infrastructure projects, and possibly

calculate the ROI of BIM implementation. However, this goal very soon had to be

reconsidered due to the following factors:

Design companies do not collect enough quantitative data on their activities,

because:

98

o Time of the consultants/designers is the most costly resource in

engineering firms

o Management and control are basic and do not employ quantitative

methods

o Companies are not willing to spend time of their employees on any

activities that do not apparently contribute to the end value, such as data

collection and measurement

Because of the time constraint, researchers who are willing to collect

quantitative data in such projects should do so without imposing burden on

designers

The effective and consistent methodology for data collection should be

implemented in projects from the beginning until the end, which is difficult

considering that a project can take several years

These barriers have shaped the current research, forcing it to re-focus on designers’

perceptions rather than on direct measurements and to use predominantly qualitative

methodology instead of quantitative. However, the experience obtained in this

research and the knowledge of the activities of design firms gained by the end of this

study made it possible to formulate the requirements for overcoming those difficulties

using some of the software tools available on the market. The recommendations also

involve some of the software functionality that does not exist yet but could easily be

implemented in some software products. These recommendations were discussed and

approved by industry professionals.

Importantly, if a research methodology can be utilised by companies not only for

research purposes but for making informed management decisions based on the data,

this would increase companies’ motivation in conducting the research. Therefore, the

development of methods for quantitative analysis of design activity in infrastructure

projects can pursue two goals:

Researchers: realise the potential of software to collect the valuable research

data on design process efficiency, planning, collaboration etc., and measure the

benefits of BIM in projects

Project managers and owners: receive up-to-date information about projects

based on collected data, in order to support management actions during and

after the projects; assess the achievement of performance goals including the

success of BIM implementation

A variety of data can be collected by researchers or designers throughout the design

project, which includes some of the data that was received during the current research.

This data includes meeting notes, recordings (audio/video), schedules, reports on

project events (errors, clashes, design changes, delays etc.), description of design

alternatives and decisions made and other descriptive information about the projects.

99

However, the capabilities of manual collection of data in design projects in AEC is

limited. It is possible to utilise some of the software products for collecting a larger

variety of data, for example by automatic registration of project events and activities

that can be used for quantitative analysis. There are at least two types of software

products that can been considered for that purpose: project bank software and

automatic clash detection software, both of which can be included in BIM solutions.

One of the examples of such software products is VDC Stream developed by Vianova.

The difference between ‘traditional’ project banks and VDC Stream is that VDC

Stream is a metadata based, and not a folder based, document management system. It

can perform automatic versioning, and in addition to downloading the newest revision

the users can download any of the older versions of the data. Because of the metadata

structure, the project bank saves all actions related to a specific document which means

dates, times, persons, revisions of each document downloaded or uploaded. This gives

an opportunity to track the workflow in projects. It can be possible to detect project

events, for instance, work iterations, changes, delays, especially with the help of data

from other sources, and map these events on project timescale.

The second kind of software that could be used to collect data systematically is

automatic clash detection software. Such software can record information about the

revisions and the identified clashes, which can be used for analysis.

One of the examples is VDC Explorer by Vianova, which is in fact a model integration

software with automatic clash detection functionality embedded. The tool is used for

creating collaboration models by combining discipline-specific designs, regardless of

their native design software, through open data formats such as LandXML and IFC.

The clash detection functionality can identify the value of clashes and find the most

critical ones (often from thousands of detected clashes). There is also an embedded

tool to comment and discuss the clashes. This can allow collecting a variety of data

related to clash detection, such as dates, disciplines involved, potential costs, causes,

corrections made etc. Textual analysis algorithms and other mathematical and

statistical methods can be applied to analyse the collected data.

It is also possible to utilise software to distribute questionnaires or quizzes among

designers to study attitudes and perceptions. The format of questionnaire is unsuitable

for use in design projects on a regular basis because it takes too much time to fill in. It

is possible to distribute single questions or small quizzes from a larger set of questions

randomly among designers. This way statistical significance will be achieved without

bothering designers too much. Another more sophisticated approach is when specific

questions or quizzes are triggered by project events.

The data collection can cover the whole project, and all data can be mapped together

on the project timeline (figure 27).

100

Figure 27: Integration of various types of data on the project timeline.

The chronological analysis of events will reveal problems and causal relationships

between them. It will outline the areas where improvement is needed. The received

information can support management actions during and after projects.

The potential of software to aid data collection is very promising since software can

help overcoming the restrictions related to the cost of designers’ time mentioned

above. The practical limitations of this particular study, such as the limited timeframe

for research, did not allow studying whole ongoing projects, and forced to focus on

past case projects. However, the suggestions outlined above can be used in more

extensive studies in the future and can help measuring the design performance and the

impact of BIM on design process quantitatively.

101

References

Aapaoja, A., (2014) ‘Enhancing value creation of construction projects through early stakeholder

involvement and integration’ [Online], University of Oulu [Accessed: 25 August 2014] Available:

<http://herkules.oulu.fi/isbn9789526204628/isbn9789526204628.pdf>

Aapaoja, A., Herrala, M., Pekuri, A., and Haapasalo, H., (2013) ‘The characteristics of and

cornerstones for creating integrated teams’, International Journal of Managing Projects in Business,

6(4): 695-713

Anumba, C. J., and Kamara, J. M., (2012) ‘Concurrent Engineering in Construction’, in: Akintoye, A.,

Goulding, J., and Zawdie, G. (Eds.), Construction Innovation and Process Improvement, Oxford, UK:

Wiley-Blackwell, 277-292

Arayici, Y., et al., (2011) ‘Technology adoption in the BIM implementation for lean architectural

practice’, Automation in Construction, March 2011, 20(2): 189-195

Baldwin, A. N, Thorpe, A., and Carter, C. D., (1996) ‘The construction alliance and electronic

information exchange: A symbolic relationship’, in: Langford, D. A., and Retik, A. (Eds.),

International Symposium for the Organization and Management of Construction: Managing

Construction Information, 1996, 3: 23-32

Ballard, H. G., (2000) ‘The Last Planner System of Production Control’ [Online], Lean Construction

Institute, May 2000 [Accessed: 4 June 2014] Available:

<http://www.leanconstruction.org/media/docs/ballard2000-dissertation.pdf>

Barlish, K., and Sullivan, K., (2012) ‘How to measure the benefits of BIM – a case study approach’,

Automation in Construction, 2012, 24: 149-15

Beck, K., et al., (2001) ‘Manifesto for Agile Software Development’ [Online; published online: 2001;

accessed: 10 February 2014] Available: <http://agilemanifesto.org/>

Bergly, G., (2001) ‘Productivity and efficiency in the construction industry’, Construction Economics

and organisation, 2nd Nordic conference, Göteborg, 24-25 April 2001, Chalmers University of

Technology, 165-172

Bogus, S. M., Molenaar, K. R., and Diekmann, J. E., (2005) ‘Concurrent Engineering Approach to

Reducing Design Delivery Time’, Journal of Construction Engineering and Management, 1 November

2005, 131(11): 1179-1185

Brown, D. C., (1996) ‘Routineness revisited’, in: Waldron, M. B., and Waldron, K. J. (Eds.),

Mechanical Design: Theory and Methodology, NY: Springer, 1996, 195-208

Brown, D. C., and Chandrasekaran, B., (1985) ‘Expert Systems for a class of mechanical design

activity, in: Gero, J. S. (Ed.), Knowledge Engineering in Computer-Aided Design, 1985, 259-276

Building SMART Finland, (n.d.) ‘Building SMART Finland’ [Online, accessed: 7 November 2014]

Available: <http://www.en.buildingsmart.kotisivukone.com/>

Coates, P., et al., (2010) ‘The key performance indicators of the BIM implementation process’

[online], in: The International Conference on Computing in Civil and Building Engineering, 30 June -

2 July 2010, Nothingham, UK [Published online: 17 August 2010; accessed: 11 February 2013]

Available at University of Salford Institutional Repository: <http://usir.salford.ac.uk/9551/>

102

Coyne, R. D., Rosenman, M. A., Radford, A. D., and Gero, J. S., (1987) ‘Knowledge-based computer-

aided architectural design’, in: Gero, J. S. (Ed.), Knowledge Engineering in Computer-Aided Design,

North-Holland, Amsterdam, 57-81

Damielo, R., (2011) ‘The Basics of Process Mapping’, 2nd Ed., CRC Press

Davern, M. J., and Kauffman, R. J., (2000) ‘Discovering potential and realizing value from

information technology investments’ [online], Journal of Management Information Systems - Special

issue: Impacts of information technology investment on organizational performance, M. E. Sharpe,

Inc., Armonk, NY, USA, March 2000, 16(4): 121-143 [Abstract in the ACM Digital Library, ref. no.

1189433; accessed: 3.6.2014] Available: <http://dl.acm.org/citation.cfm?id=1189433>

De la Garza, J. M., Alcantara, P., Jr., Kapoor, M. and Ramesh, P. S., (1994) ‘Value of Concurrent

Engineering for A/E/C Industry’, Journal of Management in Engineering, May/June 1994, 10(3): 46-

55

Eastman et al., C., Teicholz, P., Sacks, R., and Liston, K., (2008) ‘BIM Handbook: A Guide to

Building Information Modeling for Owners, Managers, Designers, Engineers, and Contractors’, John

Wiley & Sons, Inc., Hoboken, NJ, USA

Eppinger, S. D., (2001) ‘Innovation at the Speed of Information’, Harward Business Review, January

2001, 149-158

Fulford, R. and Standing, C., (2014) ‘Construction industry productivity and the potential for

collaborative practice’, International Journal of Project Management 32(2): 315-326

Gallaher, M. P., O'Connor, A. C., Dettbarn, J. L., Jr., and Gilday, L. T., (2004) ‘Cost Analysis of

Inadequate Interoperability in the U.S. Capital Facilities Industry’, August 2004

Gann, D., and Salter, A., (1998) ‘Learning and innovation management in project-based, service-

enhanced firms’, International Journal of Innovation Management, 2(4): 431-454

Gero, J. S., (1990) ‘Design prototypes: a knowledge representation schema for design’, AI Magazine,

winter 1990, 11(4): 27-36

Gero, J. S., and Kannengeisser, U., (2004) ‘The situated function–behaviour–structure framework’,

Design Studies, July 2004, 25(4): 373-391

Hicks, B. J., (2007) ‘Lean information management: Understanding and eliminating waste’, International Journal of Information Management, August 2007, 27(4): 233-249

InfraBIM, (n.d.) ‘InfraBIM, Inframallintamisen Yhteistyöfoorumi’ (‘InfraBIM, Infrastructure

Modeling Cooperation Forum’ - Finnish) [Online, accessed: 7 November 2011] Available:

<http://www.infrabim.fi/>

Kang, J. H., Anderson, S. D., and Clayton, M. J., (2007) ‘Empirical Study on the Merit of Web-Based

4D Visualization in Collaborative Construction Planning and Scheduling’, Journal of Construction

Engineering and Management, 133(6): 447-461

Karlsson, K., and Åhlström, P., (1996) ‘The difficult path to lean product development’, Journal of

Product Innovation Management, July 1996, 13(4): 283-295

Khanzode, A., (2012) ‘Making the integrated Big Room better’ [Online], DPR Construction, October

2012 [Published online: 23 October 2012; accessed: 12 February 2013] Available at:

<https://www.dpr.com/media/technical-papers/making-the-integrated-big-room-better>

103

Kiviniemi, A., (2005) ‘Requirements Management Interface to Building Product Models’ [Online],

CIFE Technical Report 161, March 2005, Stanford University [Published online: 3 March 2005;

accessed: 11 February 2013] Available: <http://cife.stanford.edu/node/185>

Kleinsmann, M., (2008) ‘Barriers and enablers for creating shared understanding in collaborative

design projects’, Design Studies, July 2008, 29(4): 20-32

Kleinsmann, M., (2010) ‘Understanding the complexity of knowledge integration in collaborative new

product development teams: A case study’, Journal of Engineering and Technology Management,

2010, 27: 20-32

Krishnan, V., Eppinger, S. D., and Whitney, D. E., (1995) ‘Accelerating product development by the

exchange of preliminary product design information’, Journal of Mechanical Design, December 1995,

117: 491-498

Kvan, T., (2000) ‘Collaborative design: what is it?’ Automation in Construction, 2000, 9: 409-415

Lædre, O. and Haugen, T. I., (2001) ‘Use of project partnering in construction: examining the effect of

project integration and target pricing in three pilot projects’, Construction Economics and

organisation, 2nd Nordic conference, Chalmers University of Technology, Göteborg, 24-25 April

2001, 139-146

Larman, C., (2004) Agile and Iterative Development: A Manager's Guide, Agile software

development series, Boston: Pearson Education

Licker, J. K., Sobek, D. K., Ward, A. C., and Cristiano, J. J., (1996) ‘Involving suppliers in product

development in the United States and Japan: Evidence for set-based concurrent engineering’, IEEE

Transactions on Engineering Management, May 1996, 43(2): 165-178

Mahmood, M. A., and Mann, G. J., (1993) ‘Measuring the organizational impact of information

technology investment: an exploratory study’ [Online], Journal of Management Information Systems -

Special section: Realizing value from information technology investment, M. E. Sharpe, Inc.,

Armonk, NY, USA, June 1993, 10(1): 97-122 [Abstract in the ACM Digital Library, ref. no. 1189687;

accessed: 3.6.2014] Available: <http://dl.acm.org/citation.cfm?id=1189687>

Maier, A.M., Eckert, C.M., and Clarkson, P.J., (2005) ‘A meta-model for communication in

engineering design’, CoDesign: International Journal of CoCreation in Design and the Arts, December

2005, 1(4): 243-254

Matthews, O., and Howell, G. A., (2013) ‘Integrated project delivery an example of relational

contracting’, Lean Construction Journal, April 2005, 2(1): 46-61

Moeller, S., (2010) ‘Characteristics of services – a new approach uncovers their value’, Journal of

Services Marketing, 2010, 24(5): 359-368

Mom, M., and Hsieh, S.-H., (2012) ‘Toward performance assessment of BIM technology

implementation’ [Online], in: Telichenko, V., Volkov, A. and Bilchuk, I. (Eds.), 14th International

Conference on Computing in Civil and Building Engineering, Moscow, 27-29 June 2012 [Published

online: 17 August 2010; accessed: 11 February 2013] Available at:

<http://www.icccbe.ru/paper_long/0187paper_long.pdf>

Morgan, J. M., and Liker, J. K., (2006) ‘The Toyota product Development Systems: Integrating

people, process and technology’, Taylor & Francis, Business & Economics, 2006

104

Ohno, T., (1998), ‘Toyota Production System: Beyond Large-Scale Production’, Productivity Press,

March 1998

Parasuraman, A., Zeithaml, V. A., and Berry, L. L., (1985) ‘A conceptual model of service quality and

its implications for future research’, Journal of Marketing, Autumn 1985, 49(4): 41-50

Parrish, K., Wong, J.-M., Tommelein, I.D., and Stojadinovic, B., (2007) ‘Communication and process

simulation of set-based design for concrete reinforcement’, in: Simulation Conference, winter 2007, 9-

12 December 2007, Washington, DC, 2057-2065

Parsons, M., Singer, D., and Sauter, J., (1999) ‘A Hybrid Agent Approach For Set-Based Conceptual

Ship Design’, International conference on Computer Applications in Shipbuilding, Cambridge, 1999,

7-11

Pei, E., Campbell, I., and Evans, M. A., (2009) ‘Building a Common Ground – The Use of Design

Representation Cards for Enhancing Collaboration between Industrial Designers and Engineering

Designers’, in: Undisciplined! Design Research Society Conference 2008, Sheffield Hallam

University, Sheffield, UK, 16-19 July 2008

Plume, J., and Mitchell, J., (2007), ‘Collaborative design using a shared IFC building model -

Learning from experience’, Automation in Construction, 16: 28-36

Poerschke, U., Holland, R., Messner, J. I., and Pihlak, M., (2010) ‘BIM collaboration across six

disciplines’ [Online], in: Tizani, W. (Ed.), Proceedings of the International Conference on Computing

in Civil and Building Engineering, 2010 [Accessed: 18.11.2014] Available:

<http://www.engineering.nottingham.ac.uk/icccbe/proceedings/pdf/pf288.pdf>

Royce, W., (1970) ‘Managing the Development of Large Software Systems’, Proceedings of IEEE

WESCON, August 1970, 26: 1-9

Sattineni, A., and Bradford, R. H., II, (2011) ‘Estimating with BIM: a survey of us construction

companies’ [Online], Proceedings of the 28th ISARC, Seoul, Korea, International Association for

Automation and Robotics in Construction, 564-569 [Online proceedings database; accessed: 3.6.2014]

Available: <http://www.iaarc.org/publications/fulltext/S16-6.pdf>

Schwaber, K., and Sutherland, J., (2013) ‘The Scrum Guide. The Definitive Guide to Scrum: The

Rules of the Game’ [Online], Scrum.org, July 2013 [Accessed: 11 June 2013] Available at:

<https://www.scrum.org/Portals/0/Documents/Scrum%20Guides/2013/Scrum-Guide.pdf#zoom=100>

Singh, V., Gu, N., and Wang, X., (2011) ‘A theoretical framework of a BIM-based multi-disciplinary

collaboration platform’, Automation in Construction, 20: 134-144

Succar, B., (2009) ‘Building information modelling framework: A research and delivery foundation

for industry stakeholders’, Automation in Construction, May 2009, 18(3): 357-375

Succar, B., Sher, W., and Williams, A., (2012) ‘Measuring BIM performance: Five metrics’ [Online],

Architectural Engineering and Design Management, 2012, 8: 120-142, [Published online: 26 March

2012; accessed: 10 February 2014] Available:

<http://www.tandfonline.com/doi/abs/10.1080/17452007.2012.659506>

Wegner, D. M., (1995) ‘A Computer Network Model of Human Transactive Memory’, Social

Cognition, 13(3): 319-339

105

Weill, P., (1992) ‘The relationship between investment in information technology and firm

performance: A study of the valve manufacturing sector’ [Online], Information Systems Research,

December 1992, 3(4): 307-333 [Abstract, published online: 1 December1992; accessed: 3 June 2014]

Available: < http://dx.doi.org/10.1287/isre.3.4.307>

Wilson, T., Wiebe, J., and Hoffmann, P., (2008) ‘Recognizing contextual polarity: An exploration of

features for phrase-level sentiment analysis’ [Online], Computational Linguistics, 35(3) [Accessed:

16.6.2014] Available: <http://www.mitpressjournals.org/doi/pdf/10.1162/coli.08-012-R1-06-90>

Womack, J. P., Jones, D. T., (2003) ‘Lean Thinking: Banish Waste and Create Wealth in Your

Corporation’, NY: Free Press

- 1 -

Appendix: The questionnaire form

Collaborative design practices and BIM: survey for project teams in infrastructure sector

In: "Measuring the benefits of Building Information Modelling (BIM) and process improvement for collaborative design in infrastructure projects" Alexander Gerbov, Vishal Singh, Maila Suvanto and Heikki Halttula,

Aalto University and Vianova Systems Finland Oy

About the survey:

In this survey you will be asked about your professional background and your experiences

of collaboration and software use. It will prepare the ground for our workshop discussion.

The survey will include 11 questions and will last about 10 minutes.

Note: the information provided by you for the research is confidential. Personal

identifications or information that might constitute personal or commercial secret may be

asked only for research purposes, and will never be disclosed. The published results will be

detached from any information that could help to connect them with personal or companies’

identities.

1. How much do you know about BIM?

Please, circle the correct answer (1-5) from the relevant column

Example: 1.

I have never worked with BIM and…

1. I have never heard about BIM

2. I have little knowledge about BIM

3. I have general knowledge about BIM

4. I have somewhat good knowledge

about BIM

5. I have good knowledge about BIM

I have worked with BIM and…

1. I have little knowledge about BIM

2. I have general knowledge about BIM

3. I have somewhat good knowledge

about BIM

4. I have good knowledge about BIM

5. I have deep knowledge about BIM

2. How much do you want to work in projects that use BIM in future?

I don’t want to 1 2 3 4 5 6 7 I want very much

3. How much do you agree with the statement: “using BIM gives clear and strong benefits

to designers and engineers”?

- 2 -

Do not agree 1 2 3 4 5 6 7 Strongly agree

0 - I am not familiar with the benefits of BIM

4. According to your understanding, what are the most important features of BIM?

Please, name 1-3 most important features of BIM that you know, or have heard of

5. How well do you know about the features of BIM, in general?

Very little 1 2 3 4 5 6 7 Very well

0 - I do not know anything about features of BIM

6. What are the problems that you face most often when working with other designers

and engineers?

Please, name 1-3 most usual problems related to working with other designers and

engineers

7. In your opinion, how efficient is the way designers work in the infrastructure projects?

Very inefficient 1 2 3 4 5 6 7 Very efficient

- 3 -

8. How satisfied are you with the communication in the project teams that you work in?

Not satisfied 1 2 3 4 5 6 7 Very satisfied

9. According to what you know, please rate the effect that BIM has on the following

parameters:

Circle the value that, in your opinion, represents how BIM affects the following

parameters.

N – I do not know at all /never heard

-3 – Very negative

3 – Very positive

a. Awareness of project's state and

actions of other designers N - 3 - 2 -1 0 1 2 3

b. Searching and finding the right

data N - 3 - 2 -1 0 1 2 3

c. Ease and joy of working N - 3 - 2 -1 0 1 2 3

d. Project delivery time N - 3 - 2 -1 0 1 2 3

e. Time required for design changes

and additions N - 3 - 2 -1 0 1 2 3

f. Collaboration and information

exchange N - 3 - 2 -1 0 1 2 3

g. Visual exploration of design N - 3 - 2 -1 0 1 2 3

h. Quality of design N - 3 - 2 -1 0 1 2 3

i. Simulation, calculation and analysis N - 3 - 2 -1 0 1 2 3

j. Clash and error detection N - 3 - 2 -1 0 1 2 3

- 4 -

10. Your educational degree

a. Bachelor b. Master c. Licentiate d. Ph.D.

e. Other, please specify: __________________________________________________

11. How many years have you worked in your field?

a. 0 - 1 b. 1 - 3 c. 3 - 5 d. 5 - 10 e. 10 - 20 f. > 20

Thank you!